14

Index Numbers
The averages, we studied under measures of central tendencies,
are used to find out a representative value of a distribution. In the next
chapter, i.e., in dispersion, we studicd the average of dispersion of
scatteredness from the central value. Here we have to deal with the
average of changes in a group of related variables, that may relate to
periods of time or between places. Generally, the changes are studied
in respect of time rather than place.
If we poSsess rupees ten, we can purchase goods or services worth
rupees ten. But it is not always the same, because the purchasing power
of money, iie., value of money changes now and then. Sometimes money
has greater purchasing power or lesser purchasing power at different
periods. The changes of money value can not be measured directly. It
depends upon the price-level. The value of money, i.e., its decrease or
increase is measured on the basis of price, i.e., its incre ase or decrease.
(AnIndex Number is a number which is used to measure the level
of a certain phenomenon as compared to the level of the same
phenomenon at some stendard period. An Index Number is a statistical
device for comparing the general lével of magnitude of a group of
related variables in two or more situations. If we want to compare the
price level of 1996 with what it was in 993, we may have to look into
agroup of variables-prices of rice, wheat, vegetables clothes, rent etc.
If the changes åre in the same ratioand in the same direction, we face
no difficulty to find out the general price-level. But practically, if we
think changes in different variables are different and that too, upward
or downward, then the price is quoted in
different units millk per litre, rice
or wheat per kilogram, rent per square feet, etc.
We want one figure to indicate the changes of different
commodities as a whole. This is called an Index Number. Index
Number is a number which indicates the changes in magnitude. John I.
Riffin says, An index number is a quantity which, by reference to a
base period, shows by its variations, the changes in the magnitude over
a period of time. In general, index numbers are used to
measure
changes over time in magnitudes which are not capable of direct
measurement." Further, A.M, Tuttle says, Index number is a single
ratio*(Usually in percentage) which measure the combined (i.e.,
A77
 478 Statistics
averaged) change of several variables between two different times
places of situations'". Again, Horace Secrist defines, "Index numbers
are serics of numbers by which changes in the magnitude of a
phenomenon are measured from time to time or from place to place»
The above discussion clearly shows that index numbers are a
measure of the average changes in a group of related variables. Index
numbers are specialiscd type of rates, percentages which give the
general level of magnitude of groups of distinct but related variables
in two or more situations. Further, .the prices of some of the
commodities may be increasing while those of others may be
decreasing during the two periods and the rates of increase or decrease
may be different for different commodities. Index number is a
statistical device which enables us to arrive at a single representalive
figure which gives the general level of the price of the phenomenon
(commodities) in an extensive groups.
Some more definitions of index numbers :
"It is anumerical value characterising the change in complex
economic phenomena over a period of time or space."
Maslow
"An index number is a statistical measure designed to show
changes in a variable or a group of related variables with respect to
time, geographical location or other characteristics. »
Spiegal
Index number are devices for measuring differences in the
magnitude of a group of related variables."
Croxton and Cowden
"An index number is a percentage relative that compares
economic measures in a given period with those same measures at a
fixed period in the past."
Clark and Schkade
"In its simplest form, an index number is nothing more than a
relative number, or a relative which expresses are relationship between
two figures, where one of the figures is used as a base."
Morris Hamburg.
Characteristics of Index Numbers
On the basis of the study and
the following points are worth analysis of definitions of Index Numbers,
1 considering:
Index Numbers are specialised
be used to compare only averages : Normally, averages can
those
the same units. But index variables which are expressed in
changes in variables which arenumbers help in
in different unifs. comparing the
 479
Index Numbers
2. Index Numbers are expressed in percentages : Index numbers are
of
expressed in terms of percentages so as to show the extent
change. However, percentage sign (%) is never used.
of direct
3. Index Numbers measure changes not capable variation in the
measurement. Where it is difficult to measure the
where the variations are
effects of agroup of variables directly orstudy,
entirely incapable of direct quantitative relative variations
are measured with the help of index
numbers.
numbers by their
4, Index Numbers are for compaison : The index
taking place
nature are comparative. They compare changes
categories.
over time or between places and like

Uses
numbers are today one of the most widelylused statistical
Index relative changes. In
devices. They are particularly useful in measuring
brief, the uses of indexnumbers are shown below :
Index number are particulary
1. They measure the relative change. idea of changes
They give better
useful in measuring relative changes.
business activity, employment, etc.
in levels of prices, production,
numbers are used to measure the
Further, Bowley observes, Index
some quantity, which we cannot observe directly."
changes to
The index numbers reduce the
2. They are of better comparison. and understandable form. The
changes of price level into more useful to percentages which are
numbers of the changes are further reduced
easily comparable.
numbers are not restricted to the
3.. They are good guides. Index phenomenon, which is spread over a
price phenomenon alone. Any expressed numerically through index
period to time, is capable of beingindex numbers serve different uses.
numbers. Thus, various kinds of Various index numbers
are Economic barometers.
4. They employment, trade, transport,
computed for different purposes, say immense value in dealing with
agriculture, industry, etc., are of
index numbers are the economic
different economic problens. Thus
barometers. their
They are the pulse of the economy. The stability of prices or
5.
deflating conditions can well be observed with the help of
inflating or general price level will measure
the
indices. Index numbers of
purchasing power of money. the wage
the wage adjuster. In all the fields of economy,
price index
6. They are the study of consumers
adjustments are done with
allowances of the employees, of the public and
numbers. Dearness determined on the basis of consumers' price
private enterprises are
index number..
 480 Statistics
7. They compare the standard of living. Different places have
different index numbers. IndexX numbers may measure the cost of living
of different classes of people. Thus comparison becomes easy in
respect of general price index numbers.
8. They are a special type of averages. Althe basiC ideas of average
are employed for the construction of index numbers. In' average, the
data are homogeneous (in the same units) : but in index number
averages the variables have different units of measurement. Hence it is
a special type.
9. They help in forulating policies. Formulation of good policies
for the future depends upon past trends. Behaviour of the index
numbers are studied carefully before making any policies. For instance,
increase or decrease in wages required to study the cost of living index
number.
10. They measure the purchasing power of money: Index numbers are
helpful in finding out the intrinsic worth of money as contrasted with its
nominal worth. Generally statements are seen that the purchasing power
of Indian rupee in 1997 is only 10 paise as compared to its
purchasing power
in 1990. It means that a person who was having an income of Rs. 100 per
month in 1990 should have an income of Rs. 1,000 to maintain the same
standard which he was maintaining in 1990. This helps in determining the
wage policy of a country.
Formulation of suitable policies in the future depends upon a careful
study of past trends. This can be done only when the bahaviour of the index
numbers during the past is examined closely. For instance, a study of the
cost of living index numbers is essential to determine whether the wages or
dearness allowance in the future have to be increased or not. Indexnumbers
are very useful in studying the trend or tendency of a series spread over a
period of time. With the help of index numbers, it is easy to find out the
trend of exports, balance of payments, industrial production, national
income, prices etc.
Types of Index Numbers
There are various types of index numbers, but in brief we shall
take three kinds, and they are :
(a) Price Index. For measuring the value of money, the general
price index is used. It is an index number which compares the
for a group of commodities at a certain time or at a place with prices
prices
Index Numbers 481

of a base period. There are wholesale price index numbers and retail
price index numbers. The wholesale price index reveals the changes in
the general price level of a country. Retail price index reveals thc
changes in the retail prices of commodities, such as, consumption
goods, bank deposits, bonds, etc.
(6) Quantity Index. Quantity index numbers study the changes in
the volume of goods produced or consumed ; for instance, industrial
production, agricultural production, import, export, ete. They are
useful and helpful to study the output in an economy.
(c) Value Index. These index numbers compare the total value of
a certain period with the total value of the base period. Here the total
value is equal to the price of each, multiplied by the quantity;for
instance, indices of profits, sales, inventories, etc.
Interpretation of Index Numbers
Index numbers are pure numbers without the units of
measurement. Suppose we say the profit for this year is 120% of the
profit for the previous year. It means ; the previous year's profit was
say Rs. 100; this year there is an increase of 20% (Rs. 20). Thus it is
clear that the profit of the current year (this year) is Rs. 120.
PROBLEMS IN THE CONSTRUCTION OF INDEX NUMBERS
The following are some guidelines to be remembered when we
construct index numbers.
1. Purpose or Object. The statistician must clearly determine the
purpose for which the index numbers are to be constructed, because
there is no all purpose index numbers. Every index number has got its
own uses and limitations. For example, if we want to study the changes
in the value of money, then we have to construct index numbers of
wholesale price. If we want to study the changes in the cost of living of
workers of any place, the cost of living index numbers of the workers
must be constructed. Cost of living index numbers of workers in an
industrial area and those of the workers of an agricultural area are
different in respect of requirements. Therefore, it is very essential to
define clearly the purposes of the index numbers and that to0
beforehand.
2. Selection of base. The base period of an index number is very
important as it is used for the construction of index numbers. Every
index number must have abase. One cannot say whether the price level
has increased or decreased, unless one compares the price level of the
current year with the price level of the previous year.
Thus when selecting a base period, the year must be recent and
normal. A normal year is one which is free from economic and natural
disturbances, widespread failure of rains, earthquakes, war, strikes,
 482 Statistics
production crisis, etc. If such abnormal years are considered, then the
index number will be a misleading onc.
Therefore, the year to be selected as base year must be normal
year or a typicalycar and a recent year. In the fast changing world of
today, the more recent the year the more representative willit be. The
base may be of the following type.
(a) Fixed base. The name reveals that the base year is a fixed one.
The prices of a particular year, selected as a base period are treated as
equal to 100. The changes in the prices of subsequent years are shown
as the percentages of the base year.
(b) Average base. Sometimes it is difficult to select an year as base
through normality. Under such a critical position, the average of
several years is considered better, as abnormalities can be reduced to
great extent.

(c) Chain base. In fixed base method, the base year once selected,
remains fixed and all index numbers are based on the same base year.
In this method, there is no fixed base year. It changes from year to year.
When a comparison is desired from year to year, a system of chain base
is used. It is the previous year that is taken as the base for the current
year ; and the change is calculated as a percentage of that year. For
instance, for 1983, the base year is 1982; for 1982, the base year is
1981, for 1981, the base year is 1980 and so on. There is no question of
normal year. Long period comparison cannot be made.
3. Selection of commodities (Selection of Regiman) If we study the
price changes of one commodity, we have to include only one item. For
instance, if we study the changes in production of cloth, then we may
include the production of mill cloth, power loom cloth, handloom
cloth, silk, khadi, etc., and there is no problem. Another exampie, say
index of retail price ; we cannot include al commodities sold in retail.
We include only the important commodities which are representative
of tastes, habits and customs of the people. For the purpose of finding
the cost of living index number, of low income group, we have to select
only those items or commodities, which are mostly consumed by that
group.
4.Source of data. The price relating to the thing to be measured
must be collected. If we want to study the changes in industrial
production, we must collect the prices relating to the production of
various goods of factories. The price may be collected from reliable
sources. The prices of commodities are the raw materials for the
construction of index numbers. The prices may be collected from the
public sources or from standard commercial magazines. The
collection of data, i.e., price must be representative,comparable and
accurate.
 483
Jndex Numbers
S. Selection of Averages. One can use any
average. But In
practice, the arithmetic average is used, because it is easy for
are difficult to
compuation ; geometric mean and harmonic mean.
calculate. But geometric mean is preferred because of the(b)following
It gives
characteristics (a) Geometric mean is the best measure and
items.
less weight to bigger items and more weight to smaller
6.Weighting. Allcommodities are not equally important. The main
changes in the
purpose of an indexX number of prices, is to ascertain the have equal
price level. In case of simple average, all commodities will
wil! have
importance. But in actual practice, different groups of people when the
different preferences on different commodities. For instance,
halved, the
price of rice or wheat is doubled and the price of sugaris isessential has
people suffer much, because the price of rice which
been doubled ; but as regards the sugar it is not so important as rice
or wheat. Therefore, to stress the importance, the system of weightingis
is adopted. John Griffin observes, "In simple terms weighting to
designed to give component series an importancc in proper relation
their real significance."
The relative importance is the basis of weightage. Generally, the
quantities of the commodities produced or the values of quantities of
the goods sold, demanded or purchased are taken as suitable weights.
These weights relate to the base year or the current years, depending
on their availability or suitability. Moreover, it is important that the
weighting system should be devised in such amanner, that it will
properly represent the commodities.
There are two methods of weighting ;
(a) Implicit weighting,
(b) Explicit weighting.
(a) Implicit weighting. Weights are not expressly laid down. The
weights are implied by the nature of commodities selected. It means
the weights relate to the selection of commodities themselves.
Emphasis is to be given to the commodities according to the number of
times the given commodities are included in the selection. For example,
a particular commodity is included say, 4 times, the weight given to the
commodity is four.
(b) Explicit weighting. Weights are expressly laid down on the basis
of importance of the items. In case only one variety is included in the
construction of index numbers, then the price is multiplied by the
quantity.
Choice of formulae

The problem before the statistician is to choose the formula which

is most appropriate and to decide about a suitable system of weights.
It is so because a large number of formulae have been devised for
 484 Statistics
constructing the index. Different formulas usually produce different
results when applicd to the same price and quantity data. Thús the
problem is perhaps of greatcr theoretical than practical importance. In
general, choice of the formula to be used depends upon the data
availablc and the nature, purpose and scope of the enquiry.
Notations
Base Year : The year selectea for which comparison i.e. the year
with reference to which comparisons are made. It is denoted by*0.
Current Year: The year for which comparisons are sought or
required.
P, = Price of a commodity in the Base Ycar.
P, = Price of a commodity in the Current Year.
q, = Quantity of a commodity consumed or purchased during
the Base Year.
q, = Quantity of a commodity consumed or purchased in the
Current Year.
W = Weight assigned to a commodity according to its relative
importance in the group.
P, = Price Index Number for the base year with reference to
the base year.
P = Price. Index Number for the current year with| reference
to the current year.
qo = Quantity Index Nunmber for the current year with reference
to the base year.
910 Quantity Index Number for the base year with reference to
the current year.
The various methods of construction of index numbers are given
below :
Methods
Index Numbers

Unweighted Weighted
(simple)

Weighted Weighted average

Simple Simple average Aggregate of price relative
Aggregate of price relative
Index Numbers 485

Unweighted
1, Simple aggregate method. This is the simplest method ol
constrùcting the index numbers. The prices of the different
commodities of the current year are added and the total is divided by
the sum of the prices of the base year commodity and multiplied by
100 : symbolically,
ZP, × 100
P, - EP,
P,
01 = Price index number for the current year with reference
to the base year.
Aggregate of prices for the current year.
2P, = Aggregate of prices for the base year.
Steps :
1. Prices of allcommodities of current years are added to get
P,
2. Prices of all commodities of base years are added to get
2P
3. EP, is divided by ZP, and multiplied with 100.
Illustration 1

Commodiy Price in 1990 Price in 1991

A 90 95
B 40 60
90 110
D 30 35

Construct an index number for 1991 taking 1990 as base.

Solution: Construction of Price Index

Commodity Price in 1990 Price in 1991

A 90 95
B 40 60

90 110
D 30 35

2 P0=250 EP=300

Pon 2P 300
X 100 = 120
250
 486, Statistics
This means that the prices have increased in the year 1991 to éhe
extent of 20% when compared with the prices of 1990.
Defects in Simple Aggregate Method
Largely numbered figures largely influence the index aumber. For
instan ce, 10% decrease in commodity A will bring down the index
number. But 10% decrease in commodity D has little influehce. The
relative importance of various commodities is not taken into account in
the index as it is unweighted.
2. Simple average of pricerelative method. In this method, the price
relative of each item is calculated separately and then áveraged. A
price relative is the price of the current year expressed as a percentage
of the price of the base year : symbolically,
|P, x 100)

Po N N
N= Number of items
When the geometric mean is employed, instead of the arithmetic
mean then the formula is
(P, x 100)
£ log
Po
Po1. = antilog
N
E logP
antilog N
P × 100
where P
P
Illustration 2
Compute a price index for the following by a (a) simple aggregate
and (b)average of price relative method by using both arithmetic mean
and geometric mean :
A B C D E
Commodity
Price in 1986 (Rs.) 20 30 10 25 40 50

Price in 1991 (Rs.) 25 30 15 35 45 55

(B.Com.Kerala)
Solution :

(a) Simple Aggregative Index = P 100

Spo = 175, Ep, = 205
ndex Numbers 487

205
:X100
175
= 117143
Calculation for Price Index
Commodiy Price in 1986. Price in 1991 Price relative log p
Po -plpr*p)x100
A 20 25 125 2.0969
B 30 30 100 2.0000

10 15 150 2.1761
D 25 3S 140 2.1461
40 45 1125 2.0511
F SS 110 2.0414
175 205 7375 125116
(b) (0) Arithmetic Mean of Price
Relatives = Po1
N
>po, = 737.5, N'= 6
737.5
6
= 12292
(ü) Geometric Mean of Price
Relative Index = Antilog log2
N

= Antilog 125116|
6
= Antilog 2.0853
= 121,7.
Merits of simple Average of Price
1. It gives equal
Relative aethod :
importance to all items.
2. Extreme items do not
unduly affect the index numbers.
3. The influence due to
diferent units is completely removed.
Demerits :
1. The use of
geometric mean involves difficulties of
computation.
2. It fails to give any
different items. consideration to the relative importance of
 Statistics
488
Weighted Index Numbers we
numbers are of two types. The index numbers
Weighted index
unweighted and they assign equal importance to
have studied above are the index. The purpose of weighting is to make
allthe items included in representative and to give more importance to
the index numbers more
them. They are :
index numbers. According to this method.
1. Weighted aggregate
weighted by quantities ; i.e., p Xq. Thus physical
prices themselves areas weights. There are various methods of assigningthe
quantities are used
various formulas have been formed for
weights, and thus
construction of index numbers.
are given below:
Some of theimportant formulae
1. Laspeyres method
2.
Paasche's method
3. Dorbish and Bowley's method
4.
Fisher's Ideal method
5.
Marshall-Edgeworth method
6. Kelly's method, and
Walsch's method
the base year quantities
7.
method. In this method,
(4) Laspeyre's
are taken as weights : symbolically,
2p,g% X 100
Po1(a) Ep,40
Steps : commodities with
year prices of various
1. Multiply the current obtain P,4o
base year weights and commodities with
year prices of various
2. Multiply the base obtain PAo
base year weights and
and multiply the quotient by 100. The
Pa,
3. Divide P,g, by us the price index.
above steps give current year quantities
method, the
(B) Paasche's method. In this
are taken as weights : symbolically,
Ep41 x100
Pos(pa)p,41
Steps : with
Multiply current year prices of various commodities
1.
current year weights and obtain P, q,.
the
Multiply the base year prices of various commodities with
2.
current year weights and obtain P, g,
Yndex Numbers 489

3. Divide P,q, by P4, and multiply the quotient by 100.

LA`PEYRE'S VS. PAASCHE'S METHODS
Laspeyre's price index is based on the assumption that the
quantities consumed in the base year and the current year are the
same. The influence of price changes on quantity demanded would not
get reflected in index numbers. But this would get reflected in
Paasche's method where the quantities of the current year are used as
weights. The two indices give different results when applied to the
same data. However, if the prices of all the goods change in the same
ratio, the two indices will be equal or if the quantities of all the goods
change in the same ratio, they willbe equal. The relationship between
the two indices depends on the correlation between the price and
quantity movements.
Laspeyres index is often preferred to Paasche's for the simple
reason that the Laspeyres index weight (go) are the base year quantities
and do not change from one year to the next. On the other hand, the
use of Paasche index requires the continuous use of new quantity
weights for each period considered. Laspeyres index is expected to
have an upward bias as it over estimates the true value. Accordingly,
Paasche's index has a downward bias and is expected to under estimate
the true value.
(C)Bowley Dorfish method. This is an index number got by the
arithmetic mean of Laspeyre's and Paasche's methods ; symbollically
(This method takes into account both | the current and the base
periods). Symbolically
}pi4% }p,41
Po(®) {po40 Ep,41 L+P
X 100 =
2
L =Laspeyre's method
P= Paasche's method
(D) Fisher's Ideal Method. Fisher's price index number is
the geometric mean of Laspeyre's and Paasche's formula given by
symbolically,
l'o() = VLXP = }p,40 Zp, 41 X 100
It is known as Ideal Index Ep,4o Epo41
Number, because;
(a) The best average, i.e., geometric mean is used for
computation.
(b) It is based on the current year as well as the
base year.
(c) It conforms certain tests of consistency.
(d) The upward bias of Laspeyre's index and downward bias of
Paasche's index are halanced to a great extent.
 Statistics
490
Method
(E) MarshallEdgewonth meau of base
Marshall-Edgeworth's index number, the arithmetic
In taken as the weights, i.e.,
year quantities is
year and current
2
Ep(4tq) x 100
Po(ta) Ep4,t4)
brackets,
By removal of
Pe =
Epo }p1i x100
EpA ~p, 41
(F Kethey's Method
period (which is neither
Kelly's index number usesquantities of some
as weights. This weight is kept
the base period nor the current period) Kelly's index number is
q, then
constant for all periods. If we denote it by
given by
Ep4 X 100
Pot*) EpA
where 9= 2

(G)' Walsch's Method:

mean of the base year and
Walsch's index nunber uses the geometric
current year quantities as weights, ie.,
w= V4%91
X 100
Then, Po (Wa) =
Illustration 3
following data :
Calculate index nunber from the
Current year
Base year
Kilo Rate (Rs.)
Kilo Rate (Rs.)
3.25
10 3
Bread
15 20
20 15
Meat
3 23
Tea 2 25
Index Numbers 491
Solution:
Construction of Price Index

Base year Current year

Kilo Rate Kilo Rate P190 POg0 Pigi Poq1
(Rs.) (Rs.)
q0 Po q1 P1
Bread 10 3 3.25 32.50 30 26 24
Meat 20 15 15 20.00 400.00 300 300 225
Tea 2 25 3 23.00 46.00 50 69 75
099
Total478.50 380 395 324

(a) Laspeyre's method

2p, q% X 100 478.50× 100
= 125,9
Epo4o 380.00
(b) Paasche's method
Po = }p,41 X 100=
395 × 100
= 121.9
}p,41 324
(c) Bowley's method
Zp40, ZP,41
2 X 100

478.5,395
380.0 324 X 100 = 123.9
2

or
L+P 125.9+121.9
2 2
= 123.9
(d) Fisher's ideal formula

X Ep,41 X 100
V478.5 X 395 Ep,41
380 324 X 100
= V1.259 X1.219 x 100
= 1.239 × 100 = 123.9
(e) Marshall Edgeworth method
Pa,
Z 01 = Ep,46+2p191
Zp,q6+2p,41
X 100 = 478.5+395
380.0+324
X 100

873.5 x 100
= 1.24 X 100 = 124.
704
 Statistics
492

average of price relative. Price relative is not

2. Weighted method. If we know thevalues
calculated by the weighted aggregate
year, then we can construct the weighted index
consumed in the base relative method. For
average of
number according to the weighted may, use the jarithmetic, mean or
averaging weighted price relatives, we
symbolically :
the geometric mean ;
P
X po 40
x 100 =
Po =
P
P = Price relative x100
Po
V= Value weight (Po 9)
Steps :
1. Find the price relatives for each commodity ie.
P X 100
p P,
corresponding weight
2. Multiply the price relatives in step 1 by
to get PV.
obtained in step2 for all the
$3. Obtain the sum of products
commodities to get 2 PV.
4. Divide the sum in step 3 by V
index based on weighted
The resulting figure gives the price
average of the relatives.
when G.M. is applied, the formula is
2Po4,log, x 100 ZVlog P
Pos 2Po40
The Steps involved are :
P,
P=x 100 for each commodity.
1. Compute price relatives i.e.
relatives. This gives log P.
2. Find the logarithms of allthe price
each commodity by the
3. Multiply log P valued, for (log. F) V.
corresponding weights, to get
4. Find the sum of the values, to get 2 (log. P) V.
5. Divide the sum obtained (in step 4) by ZV.
required price
6. Antilog of the value obtained in step 5 gives
index.
 Index Numbers
493,
Quantity Index Numbers
Price index numbers measure and permit
of certain goods ; the quantity index numbers comparison of the price
physical quantity of goods produced, consumedpermit comparison of the
or
most common type of the quantity index is that of distributed. The
quantity produced.
Laspeyre's method:2o1 Eq,P, X 100
Paasche's method:o= X 100
Eq,P,
Fisher's method : Eq, P,2q, P, X 100
Eq,P, Eq,P,
Illustration 4

Compute price index for the following

average of price relative method, using data by applying weighted
) Arithmetic and (ii)
Geometric mean.
Iiem Price in 1991 Price in 1992
Rs.
Quantiy in 1991
Rs.
Wheat 2.00 2.50 40 kg
Sugar 3.00 3.25 20 kg
Milk 1.50 1.75 10 lit.
Solution : Construction of Index
Item Price in Price in Duantiyi
1991 1992 n 1991
Poqo P= Po
PV
Log P Log P
Po P1
X 100
Wheat 2.00 2.50 40 125 10000 2.0969 167.7520
Sugar 3.00 3.25 20 60 108.33 6499.8 2.0346 122.0760
Milk 1.50 1.75 10 15 116.67 1750.05 2.0669 31.0065
Total 155 2 PV= E logPV=
18249.85 320.8345
(a) Arithmetic mean
£ PV
Po =

2 PV = 18249.85
155
117.74
 Statistics
194

(b) Geometric mean

log PV
Po,= Antilog
E log PV= 320.8345
2V= 155
320.8345
E Po = Antilog 155
= Antilog 2.0699
= 117.4
Iliustration 5
(iü) Paasche's
Calculate quantity index by (i) Laspeyre's method,
method, (iüi) Fisher's method.
1989 Price 1991
Year Price Total value
Total Value Pl
Commodiny po
(Pq1)
(poqo)
12 144
10 100

144 14 196
B 12
16 256
C 14 196
18 324
D 16 256
20 400
18 324

Construction of Various Indices

Solution.
pogi piq1 P140
90 P1 q1 poqo
po 144 120
100 120
10 12 12
10 168
168 196
14 144
12 12 14
256 224
196 224
14 16 16
14 288
288 324
18
18 256
16 16
400 360
20 324 360
18 18 20
1,320 1,160
Total 1,020 1,160
1160
L=
}p4o X 100=
1020
X 100
EpAo
= 113.7
1320 X 100
P= Ep41
X 100 =
1160
= 113.8
-F= VL xP= 113.7 x 1138
= 113.75
Index Numbers 495

Illustration :6
From the following data calculate price index numbers for 1995 with
1985 as base by (1) Laspeyre's method, (2) Paasche's method, (3)
Marshall·Edgeworth method and (4) Fisher's Ideal method :
Comnodiy 1985 1995
Price Quantiy Price uantiy
A 20 40 6
B 50 10 60

40 15 S0 15
D 20 20 20 25

(B. Com. (Hons), Delhi)

Solution:
Computation of Index Numbers
Commo 1985 1995
dity P40 Po90 P41 Po41
Po P 91
A 20 8 40 6 320 160 240 120
B S0 10 60 600 S00 300 250
40 15 S0 15 750 600 750 600
D 20 20 20 25 400 400 S00 500
|2p190 =
2070 = 1660 1790 1470

(1) Laspayre's Index = Po. X 100

2070
X 100 = 124.70
1660

(2) Paasche's Index =o, SP41 X 100

1790
X 100== 121.77
1470
(3) Marshal Edgeworth Index :
Po1 =
EP (90 + 4) + }p,91
X 1000 =
EPol4o +41) = 100
496 Statistics
2070 + 1790 3860
X 100 = x 100
1660 + 1470 3130
= 123.32
(4) Fisher's Ideal Index:

X X 100
Po1=

=
V2070 1790 X 100
1660 1470

= V1.5184 x100= 1.2322 × 100

= 123.22

Value index numbers

is the
Value index numbers are easy to calculate. Here, value of
or V is the sum the
product of price and quantity. The value index values
values of a given year divided by the sum of the of the base year.
The formula is :
The value index is a type not in wide use.
Value Index or V= 2P41 X 100 or
V= Total values of all commodities in the given
period.
the base
V, =Total values of all commodities in
period.
Test of Consistency of Index Numbers
the construction of index
Severa! formulae have been studied for
is appropriate to a
numbers. The question arises as to which formula
have been developed and the
given problem. A number of tests
important among them are:
1/Time Reversal Test
that an index number
Reversibility is an important property satisfy
good index number should the time reversal
should possess. Á
formula for calculating an
tests. In the words of Irving Fisher, "The same ratio between one
index number should be such that it gives thewhich of the two is taken
matter
point of comparison and the other, no
Index Numbers 491
as the base : or putting it in another way, the index number reckoned
forward should be reciprocal of the one reckoned backward." One of
the advantages claimed in favour of Fisher's formula is that it makes
the index number reversible. The time reversal test shows that the
following equations hold good ; symbolically
P, x Po, = 1(omitting the factor 100 from each index number)
Fisher's ideal formula satisfies the time reversal test :

Zplo PAi
Pu= SpA EpAo
X
Epa, 2pg0
}p41 2pAi PA0
Po XPy= X X X = 1=1
EpAo ZpA1
Itshows that the time reversal test is satisfied.
2/Factor Reversal Test
Another basic test is that the formula for index number ought to
permit interchanging the prices and quantities without giving
inconsistent results i.e., the two results multiplied together should give
the true value ratio. A good index number should satisfy not only the
time reversal test, but also the factor reversal test. A good index
number should allow time reversibility, interchange of the base year
and the current year, without giving inconsistent results.

Po1 VPAo ~p41

2pAo pAi
}pAo
Poxlo=
ZpAo
X pAi X Ep41
Eplo
X Ep41
Ep}pAo
p41
Illustration 7. 2pAo
Compute Index Number, using Fishers Ideal formula and show
that it satisfies time-reversal test and factor-reversal
test.
498
Statistics
Base Year Current year
Quantiy Price Quantiy Price

10 15 12
A 12
20
B 15
20 9
C 24
16 5 14
D

(M.Com. Nagpur)
Solution :
Computation of Index Number

Commodity 40 Po 41 P1 P190 Polo P41 PoA1

10 15 12 144 120 180 150

A 12
20 75 105 100 140
B 1
20 9 216 120 180 100
C 24
5 16 5 14 70 80 70 80
D
pPo ~poPo ~poPi
=505 =425 =530 =470

Po =
VPao X pa X 100
EpAo EpAi
505 530
X X 100
425 470
= V1.188 X 1.128 x 100
V1.340 x 100 = 1.158 x 100
115.8
(i) Time-Reversal Test
Time Reversal Test is satisfied when P XP = 1
X
Po=
V470 X 425
530 505
530 470 425
505
Po XP = V 425 470 530 505
= 1.
Index Numbers 499

(iü) Factor-Reversal test :

Factor-Reversal test is satisfied when

Zpo4%
qPo X

505 530 470 530

Po x Qon = V 425 470 425
X
505
530 Ep41
425 ie }po
Po,Xo1=
EpAo
Hence the given data satisfies the time-reversal test and
factor-reversal test.
Illustration 8

Construct with the help of the table below, Fisher's Ideal Index.

1987 1991

Commodity Price Quantity Price Quanity

A 6 S0 10 56
B 2 100 2 120

60 6 60
D 10 30 12 24
E 40 12

Also prove from-the above data that the Factor Reversal and
Time Reversal tests are satisfied by Fisher's ideal formula.
(Madras B. Com. Aligarh, M. Com.)
Solution : Construction of Fisher's Ideal Index
Base Quan Curent uan
Commo- Yr.Price. Piq0 P1Q1 poq1
tity yr. tiy
dities p0 g0
Price p1 q1
A 6 50 10 56 300 S00 S60 336
2 100 2 120 200 200 240 240
60 6 60 240 360 360 240
 Statistics
500
360 288 240
12 24 300
D 10 30
480 432 288
12 36 320
E 40
Ep1qo Ep1qi } poq1
Total Zpoqo 1900 1880 1344
1360

Fisher's Ideal Index

X X 100
Po = EpAo 2PA1
19001880 X 100
13601344
V1.954 X 100
= 1.398 × 100

= 139.8
satisfied if
Factor Reversal test is
Epa1
Po X lo =
EpAo
X X X
Po Xlo =
Substituting the values
V1900 X 1880 1344 1880
X X
Po X Lo1 = 1360 1334 1360 1900
1880
1360
1880
Now
Ep41 is also equal to 1360
EpA reversal test also. Time
Fisher's Ideal Index satisfies factor
reversal test is satisfied if
Po XPo=1
X
EpA1 X EpAo
EpAo EpAi
Substituting the values from the table
1990y1880 1344
P,Po XP, = V1360
1360
1344 1880 1900

= 1
Time Reversal te[t is satisficd by Fisher's Ideal formula.
 Index Numbers
501
Unit Test
This test requires that the index number formula should be
independent of the units in which the prices or quantities of various
commodities are quoted. This test is satisfied by all the formula except
the simple aggregative index.
Circular Test :
This is another test for the adequacy of an index number. It is an
extension of time reversal test. According to this test, the index should
work in a circular fashion. For instance, an index number is
constructed for the year 1985 with the base of 1984 and another index
number for 1984 on the base of 1983, then it should be possible for us
to directly get an index number for 1985 on the base of 1983. If the
index number calculated directly does not give inconsistent value, the
circular test is said to be satisfied. Their product should be equal to 1.
Po X P2 X Po = 1
This test is satisfied only by the indexX number formula based on :
(1)Simpie geumeric mean of the price relatives (2) Kelly's fixed base
method.
Chain Base Method
The base may be fixed or changing. So far we have used the fixed
base method in various formulas. In the fixed base method, the base
remains constant throughout i.e., the relative for all the year is based on
the prices of that single year. On the other hand, in the chain base
method, the relative for each year is found out from the prices of the
immediately preceding year. Thus the base changes from year to year.
The indices which we find out by this method are called
index numbers or link relatives. These link relatives link relative
are linked
together. For example, if we are to compute an index
1976, the base year is 1975 ; an index number tor 1977, the number for
is 1976;for 1978, the base year is 1977 base year
and so on.
The following formula is used for finding out the
chain index :
Linkrelative of the current year x Previous
Chain Index year Chain Index
100
Construction of Chain Indices
Steps : 1. Link relatives are found out by stating the
cach year as percentage of the preceding year; i.e., the link figures foris
relative
equal to Current year's price
Previous year's price X 100.
2. Link together these percentages by successive multiplication to
get link relatives.
 Statistics
502
current year
3. Chain index for index ofpreviousyear
Average link relative x Chain
of thecurrent year
100

Illustration 9
base index numbers into chain base
Convert the following fixed
index numbers: 1990 1991
1988 1989
1986 1987
Year: 392 400
.408 380
376 392
E.BI. :
(B.Com. Punjab,)
Solution :
Conversion of F.B.I. to C.B.I.

Link Relatives Chain Index

FBI.
Year
376
1986 376
376 × 104.26
392 392 X 100=104.26 = 392
1987 100
376
392 X104.08
=408
1988 408
4Ux100=104.08
392
100

408 x 93.14 =380

380 380 x 100=93.14
1989 100
408
380 × 103.16
392 392 X 100=103.16 =392
1990 100
380
392 X 102.04
400 400 x100=102.04 -=400
1991 100
392

Merits :
great advantage to economists and
1. The chain base method is a change that has come
businessmen, It helps them to know the extent of
year.
in the current year, as compared to the previous
2. We can introduce new items and
omit old items without any
price index and
recalculation. It is flexible and so used in consumer
wholesale price index.
than by the other
3. It is more free from seasonal variation
methods.
4. Weights can be adjusted as frequently as possible.
Index Numbers 503

Demerits :
1. It is a very difficult task toselect the items to be included or
omitted.
2. If there is any abnormal year or any defect in any place, it will
affect the subsequent year also.
Differences between Chain Base method and Fixed Base method :

Chain Base Ficed Base

1. The base year changes. The base year does not change.
No such link relative method is used.
2. Here the link relative method is used.
in the commodities, will involve
3. Introduction and eletion of items are easy Any changeindex number to be recast.
entire
to calculate, without recalculation of thethe
entire series.
The calculations are simple.
4. The calculations are tedious.
5. It is difficult to understand. It is simple to understand
is no such problem
6. It cannot be computed if data for any oneThere
year are missing.
It is suitable for long periods only.
7. It is suitable for short period only.
adjusted so frequenty.
|8. Weights can be adjusted as frequently as Weights cannot be
possible.
not so, the error is confined to the
9. Index number is wrong if an error isThis is
committed in the calculation of any link index index of that year only.
number.

Conversion of Chain index into Fixed Index

Sometimes it is necessàry to convert the chain base index into
fixed base index number. In that case the following steps must be
followed :
1. The first year is fixed as the chain base index and it willbe taken
as the base which is 100.
2. For finding out the indices for other years, the following
formula is used :
Current year's F.B.I.
Current year's C.B.1. X Previous year's F.B.I.
100
F.B.I. = Fixed Base Index
C.B.I, = Chain Base Index.
Illustration 10
From the chain index numbers given below prepare fixed base
index numbers (figures arbitrary)
 S04 Statistics
Years 1987 1988 1989 1990 1991
Chain base index Nos 80 110 120 90 140

(B.Com. Madras)
Solution :
Construction of Fixed Base Index

Year Chain Chain Base Index Numbers changed Fixed Base

to 1987 as base
base index Index
numbers Numbers
1987 80 80

1988 110 80 88
X 110
100

1989 120 80 110 105.6

X X 120
100 100

1990 90 80
X 110, 120 X 90
95.04
110 100 100

1991 140 110 120 90 133.06

X X -× 140
100 100 100 100

Base shifting :
Sometimes it is necessary to shift the base from one period to
another for comparison purposes or for other reasons like, the previous
base becoming very old to compare. By base shifting, we can express
the series in terms of a recent period. The change of reference base
period is called shifting the base. The following formula must be used
in this method of base shifting.
Index Number (based on New Base Year)
Current year's old index number X 100
New base year's old index number
Illustration 11
Reconstruct the following Index Number Numbers by'shifting base to:
) 1995 and (ü) 1997
Year 1990 1991 1992 1993 1994 1995 1996 1997
Inxex Nos. 120 150 160 180 200 200 210 240
(Bombay B. Com.)
Index Numbers 505

Year Index Numbers 1995 1997

1990 120 120 120

X 100 = 60 X 100 = 50
200 240

1991 150 150 150

X 100 = 75 X 100 = 62.5
200 240

1992 160 160

X 100 = 80 160.x 100 = 66.66
200 240

1993 180 180 180 X 100 = 75

× 100 = 90
200 240

1994 200 200 200

X 100 = 100 X 100 = 83.33
200 240

1995 200 200 200

X 100 = 100 X 100 = 83.33
200 240

1996 210 210 X 1000 = 105 210

X 100 = 87.5
200 240

1997 240 240 240

X 100 = 120 × 100 = 100
200 240

Splicing two Index Number Series :

The articles which are included in an index number may become
out of fashion or go out of the market. New articles come into the
market, for which relative importance may also change. So it is
necessary to incude the articles in the index number. The old series of
index number is discontinued and we must construct a new series and
must take the year of discontinuation as the first base.
Thus we connect the new set of index with the old discontinued
one. The statistical method connects an old index number series with
a revised series in order to make the series continuous is called
splicing. The formula is :
Index No. of Current year × Old Index No of

Spliced Index No,= New base

100
year

Illustration' :12
Two sets of Indices, one with 1976 as base and the other with 1984
as base are given below :
(a) Year Index Numbers (b) year Ihdex Numberr

1976 100
1977 110
1978 120 1984 100
S06 Statistics
190 1985 105
1979
1980 300 1986 90

1981 330 1987 95

1982 360 1988 102

1983 390 1989 110

1984 400 1990 96

The Index (a) with 1976 base was discontinued in 1984. You are
required to splice the second index number (b) with 1984 base to the
first index number.
Solution:
Splicing of Index Numbers

Year Index Number Index Number Index Number

(a) (b) (b) spliced to
witlh 1976ias base with 1984las base (a) with 1976 as base
1976 100

1977 110

1978 120

1979 190

1980 300

1981 330

1982 360

1983 390

1984 400 100 400

100 X =400
100
1985 105 400
105 × =420
100
1986
90 400
90 × =360
100
1987
95 400
95 X =380
100
1988
102 400
102 × 100408
1989
110 400
110 × =440
100
1990
96 400
96 X =384
100
Inder Numbers 507
Illustration :13
Taking 1989 as base, the index numbers of wholesale prices of a
commodity aregiven below:
Year
1989 1990 1991 1992 1993 1994 1995
Index 100 120 190 200 206 230 300
Nos.
Construct a new series taking 1992 as base.
(B.Com., Madras)

Solution:
Construction of Index Number by Base Shifting
Index Numbers Inda Numbers
Year Base (1992 = 100)
Base (1989 = 100)
100 100 X 100 == 50
1989
200

120 120 X 100= 60

1990
200

190 190 X 100 = 95

1991
200

200 200
1992 X 100= 100
200
206 206 X 100 = 103
1993
200

230 230 X 100 = 115

1994
200

300 300 X 100 = 150

1995
200

Deflating Index Number

figure for changing in levels
It is useful to adjust a series of rupee allowance for the effect of
of prices. It means that we are making price level are
changing price levels. The cost of living andthethe wage level is also
increasing year by year. At the same time,
increasing. The real wage level is decreasing more than the mÍney
is decreasing or the value of
wage ; i.e. the purchasing power of money
money is decreasing ; and so the cost of living is increasing and the
The process of deflating or decreasing a
price level is also increasing. numbers, so as to allow for change in the
figure with the help of index wages
this method a series of moneytevet
price level is called deflating. Byprice of
changes to find out the
or income can be corrected for js :
real wages or income. The formula
 508

Money wage
Statistics
Real wage = Price Index X 100
Real Wage of Income Index No.
Index of money wage X 100
Price Index No.

or
Realwage of the current year X 100
Real wage of the base year
Illustration: 14
Given the following data:
Year Weekly take-home Consumner Price
pay (wages) Index
1991 109.5 112.8
1992 112.2 118.2
1993 116.4 127.4
1994 125.08 138.2
1995 135.4 143.5
1996 138.1 149.3

(1) What was the real average weekly wage for each year ?
(2) In which year did the employee have the greatest buying power?
(3) What percentage increase in the weekly wages for the year 1996
is required, if any, to provide the same buying power that the
employees enjoyed in the year in which they had the highest real
wages ?
(B.C. (Hons.), Delhi)
Calculation of Real Wages
Year Weekiy take-home pay Consumer price Index Realwages
(Rs.)
1991 109.5 112.8 109.5
X 100 = 97.07
112.8

1992 112.2 118.2 112.2

X 100 = 92.92
118.2

1993 116.4 127.4 116.4

X 100= 91.37
127.4
1994 125.08 138.2 125.08
X 100= 90.51
138.2
1995 135.4 143.5 135.4
X 100 = 94.36
143.5
1996 138.1 149.8 138.1
× 100= 92.19
149.8
Tndex Numbers 509
Solution:
(1) Real average weekly wage can be obtained by the formula :
Real Wage Money Wage X 100
Price Index
(2) The employee had the greatest buying power in 1991 as the real
wage was maximum in 1991,
(3) Absolute difference = 97.07 -92.19 = + 4.88
Consumer Price Index
Consumer Price Index is also called as the cost of living index.
Statisticians recommend that the terms 'cost of Iiving index'or 'price of
livingindex' or 'cost of living price index' or consumer price index can
be used in the appropriate place. In different countries, cost of living
index, consumer price index and retail price index are used.
Meaning and need
Consumer price index numbers are designed to measure the
average change over time in the price paid by ultimate consumer for a
specified quantity of goods and services. Consumer price indices or
cost of living indices measure the change in the cost of living of workers
due to change in the retail price. Achange in the price level affects the
cost of living of different classes of people differently. The general
index number fails to reveal this. So there is the need to construct
consumer price index. People consume different types of commodities.
People's consumption habit is:also different from man to man, place to
place and class to class, i.e., richer class, middle class and poor class.
The scope of consumer price is necessary, to specify the
population groups covered ; for example, working class, middle class,
poor class, richer class, etc., and the geographical areas must be
covered as urban, rural, city, town, etc.
Uses of consumer price index
1. This is very useful in wage negotiations and wage contracts
and allowance adjustment in many countries.
2. Government can make use of these indices for wage policy,
price policy, taxation, general economic policies and rent
control.
3. Changes in the purchasing power of money and real income
can be measured.
4. We can analyse the market price for particular kinds of goods
and services by this index.
Construction of a consumer price index
There are some precautions to be taken in the construction of
price index.
 S10 Statistics
L. Determination of the class of people, for whom the index i
required is necessary. Before constructing the consumer price
we must decide the class of people (rich or poor), and the area to index,
which the price index is related (rural or urban) must be
carefully
determined. For example, we must determine whether it relates to tha
consumption habit of the rich or poor people, who are living in urban
or rural areas. We must clearly define the scope of the index, Fo.
example, when we consider the consumption habit of the working class,
we must refer to industrial workers or agricultural workers.
2. Selection of base period :
The base period must be a period of economic stability
Fluctuations due tochange in scason may be eliminated by taking à
complete year as a base period.
3. Conducting family budget enquiry :
The family budget enquiry or expenditure survey covers the
population group for whom the index is meant. The object of it is to
determine the amount spent on different items of consumption by the
average family in a population group. We must take quantities of
commodities consumed and the price level of cach commodity. Thus
we can find out the consumption pattern. Family budget survey is based
upon a random sample basis, where families are selected by lots.
The expenditure on differcnt items is classified into broad heads
and in certain weli-defined groups. They are:
1. Food
2. Clothing
3. Fuel and lighting
4. House rent
5. Miscellaneous
various
These broad groups can be further subdivided into wheat,
sub-groups. The broad group of food may be divided into rice,
commodities which do not
pulses, sugar, etc. We must include those we
have variation in quantities or seasonal variation in supply ; and
must get a regular price quotation of the commodities.
Collection of price quotations :
Collection of price quotations is very important in the
construction of cost of living index prices which vary from place to
place, to shop and from consumer to consumer, and it is a difficult and
time consuming job tocollect retail price. The principles to be adopted
in the collection of retail prices are :
1. Price must relate to a fixed list of items for a fixed quality.
2. Retail price must be the price which is given by the consumer.
Index Numbers 'si
3, If discount is given to all customers, it can be taken into
account.,
4. We must take into account the ration price or control price and
open market price.
In the cost of living Index, price is the most important element. So
ue must be very careful in the collection of price lists. Price lists are
collected by special agents or mail questionnaires. First we must select
special agents and give special training to them. They must be instructed
verification
Droperly and given the list of items to be priced. Price
nust be conducted by 'check pricing' or 'purchase checking. index
The cost of living index is a weighted index. The price of the
different
number of its relatives must be weighted, because there are
of people. It
relative importänce of various items for different classes
items.
can be found out by the amount spent on various
Index
Method of constructing Consumer Price
Price Index.
There are two methods of constructing Consumer
They are :
method.
1. Aggregate Expenditure method or Aggregate Relatives.
2. Family Budget method or method of Weighted
Aggregate Expenditure method
This method is based upon the Laspeyre's method. It is widely
used. The quantities of commodities consumed by a particular group in
the base year are the weight. The formula is :
Consumer Price Index Number = × 100
EpAo
PRECAUTIONS IN THE USE OF COST OF LIVING INDEX
NUMBERS
Many times, the consumer price index numbers or cost of living
index numbers are misinterpreted. Therefore, are the points to be kept
in mind while using these indices.
1. Cost of living index numbers only measure changes in retail
price from one period to another period. It does not tell us anything
about variations in the living standard at different places.
2. Weights may not be representative and if it is so index numbers
would give misleading conclusions.
3. The index numbers do not take into account the changes in
qualities.
Illustration 15
Calculate the Index Number using the Aggregate Expenditure
method for the year 1993 with 1990 as base year, from the following data :
 512
Statisticy
Commodiy Quantiy in wnis Price per unit Price per unit
in 1990 (Rs.) in 1993 (Rs.)
A 100 12.00
B 25 6 7.50
C 10 S.25
D 20 48 S2.00
65 15 16.50
F 30 19 27.00

(B.Com. Bombay.)
Solution:
Calculation of Index Number by using Aggregate Expenditure
Method.
Commodiny Po P1 Piqo Poqo
A 100 12.00 1200.00 800
B 25 6 7.50 187.50 150
C 10 5.25 52.50 50
D 20 48 52,00 1040.00 960
E 65 15 16.50 1072.50 975
F 30 19 27.00 810.00 570

Total 4362.5 3505

Po1 Ep40 x 100 = 4362.50

3505
X 100
Epfo
= 124.46
Family Budget method
Here an aggregate expenditure of an average family on various
items is estimated and it is value weighted. The formula is
£ PV
Consumer Price Index

where P = Pi X 100 for each item and =value weight i.e., poqo.
Po
"Weighted average price relative method" which we have studied
before and "family budget method'" are the same for finding out
consumer price index.
Index Numbers 513

Illustration : 16,
Calculate index number of prices for 1996 on the basis of 1995
:
from the data given below
Price pe unit Price per unit
Commodity Weight
1979 1989

40 16.00 20.00
A
25 40.00 60.00
B
5 0.50 0.50

20 5.12 6.25
D
2.00 1.50
10
(Lucknow)
Solution :
Calculation of Index Numbers

Price per Price per Price Weighied re

Commodity Weights unit ; 1996 relatives P latives;P.V.
unit ; 1g95
po P1
(P
po
x 100)
40 16.00 20.00 125 S,000
A

B 25 40.00 60.00 150 3,750

0.50 0.50 100 500

C

D 20 5.12 6.25 122.1 2,442

E 10 2.00 1.50 75 750

£V=100 2PV
= 12,442

Index number of prices for 1996

EPV

12,442 = 124.42
100
Price Index Number for 1996 = Rs. 124.42
Limitations of Index Numbers
Even though index numbers are very important in business and
economic activities, they have their own limitations ; they are :
 314
Statistics
1, There may be error in each stage of the construction of the
index number, namely, selection of commodities, selection of
the base period, selection of weight, etc.
2. Index numbers may not represent the exact change in price
level, because they are based on sample data.
3. Tastes, habits and customs of people change in course of time
and may make the weighting not suitable for the present data,.
4. In cach index there is an index error, because there is no
formula for measuring the price change. So there is the
formula error. Hence it will not be a representative one.
5. By selecting a suitable year as the base year, selfish persons
may get their desired results.
Index numbers, properly constructed, will be useful as economic
baraometer.
IHustration 17
Solution: Construction or index Numbers
Year Price Index Number
Current year price x 100
Base year price
1981 40 100

1982 42 42 X 100= 105

40

1983 35 35 X 100=87.5
40

1984 33 33 X 100=82.5
40

1985 39 39
x 100=975
40

1986 40 40
X 100=100
40

1987 42 42
x 100=105
40

1988 48 48
x 100=120
40

1989 (sS x 100=137.5

40
1990 70 70 X 100=175
40
 315
Index Numbers
Rrepare Index numbers (1981 = 100)of prices of a commodity
from the following data :
1983 1984 1985 1986 1987 1988 1989 1990
Year 1981 1982

35 33 39 40 42 48 5S 70
Price 40 42

(B.Com. Andhra)

llustration 18
index numbers and
The following table gives the group price
percentages of expenditure on various groups, obtained from inquiries
families of Bombay.
into the budget of the middle class
Index Number
Percentag of Group price
Group 1989 1991
Expendinure
350 440
35
Food
220 330
Fuel and Lighting 10
230 400
Clothing 20

160 105
15
Rent
190 340
Miscellaneous 20

comparea
What changes are in the level of cost of living of 1991 is
to that of 1989.
(BA. Bombay)
Solution : Calculation of Cost of Living Index
PxV
Group 1989 1991
P=x 100 Percenage
{Po) = V
Po

Food 350 440 440 35 4399.85

X 100=125.71
350

Fuel and 220 330 330 10 1500.00

X 100=150.00
Lighting 220

Clothing 230 400 400 20 3478.20

x 100=173.91
230

Rent 160 105 105 15 984.45

X 100=6S.63
160

Miscell 190 340 340 20 3579.01

aneous
X 100=178.95
190

EV=100 2 PV =
13,941.3:
 516
Statistics
£ PV
C.I. =

13941.50
100
= 139,415
Cost of living of 1991
= 139.42
Illustration : 19
An enquiry into the budgets of middle class families in a certain city
gave the following information :
Expenses Food Fuel Clothing Rent Miscellaneous
35% 10% 20% 15% 20%
Prices (1995) Rs. 150 Rs. 25 Rs. 75 Rs. 30 Rs. 40
Prices (1996) Rs. 145 Rs. 23 Rs. 65 Rs. 30 Rs.45
What is the cost of living index number of 1996 as compared with that
of 1995 ?
(B.Com, Kerala)
Solution:
Construction of Cost of Living Index

Expenses 1995 Po 1996 P P; x 100 P PW

Po
Food 150 145 96.67 35 3383,45
Fuel 25 23 92.00 10 920.00

Clohing 75 65 86.67 20 1733.40

Rent 30 30 100.00 15 1500.00
Miscellaneous 40 45 112.50 20 2250.00

2= 100 EPW=
9786.85

SPW
Cost of Living Index EW
9786.85
100
=97.87
0ndex Numbers 517

Illustration :20
Construct the consumer price index number for 1996 on the basis of
1995 from the foilowing data using the aggregate expenditure method:
Commodity Quantity Consumed Prices in 1995 Prices in 1996
1995 Rs. Rs.
A 6Quintal 5.75 6.00
B 6Quintal 5.00 8.00
C 1Quintal 6.00 9.00
D 6Quintal 8.00 10.00
E 4Kg 2.00 1.50
F 1Quintal 20.00 15.00
(B.Com, Bangalore)
Solution:
Construction of Consumer Price Index Number

Commodiy Quantity Units Price in Price in P140 Po 40

Consumed 1995 Po 1996 P
in 1995. 40

A 6 Qtl. Qtl. 5.75 6.00 36.00 34.50

6 Qtl. Qtl. S.00 8.00 48.00 30.00

1Qt1. Qtl. 6.00 9.00 9.00 6.00

D 6 Qtl. Qtl. 8.00 10.00 60.00 48.00

4 Kg. Kg. 2.00 1.50 6.00 8.00

F 1Qtl. Qtl. 20.00 15.00 15.00 20.00

2p190= 174
= 146.50

Consumer Price Index = Zpi40 X 100

~po 40
174
X 100
146.50
= 118.77
Illustration : 21
From the following data calculate an index number using family
budget nethod for the year 1993 with 1990 as base year.
 j18 Statistics
Lommodity uantity (1990) Price per unjt
(in units) Rs.
1990 1993
A 100 8.00, 12.00
B 25 6.00 7.50
10 500 5.25
D 20 48.00 60.00
25 15.00 16.50
F 30 9.00 27.50
(B. Com., Kamataka)
Solution:
Calculation of Index Number by family Budget Method
Commodity Quantity W Po P P; PW
X 100=P
Po
A 100 8.00 12.00 150 15,000
B 25 6.00 7.50 125 3,125
C 10 S.00 5.25 105 1,050
D 20 48.00 60.00 125 2,500
E 25 15.00 16.50 110 2,700
30 9.00 27.50 300 9,000
EW= 210 2PW = 33,425
EPW
Index Number
EW
33425
210
= 159.17

Miscellaneous Illustrations
Illustration 22
With the help of the following data prove that Fisher's Ideal Index
satisfies both the time reversal test and factor reversal test.
(M. Com. Jodhpur)
krdex Numbers 519

1980 1990
Comnodiy Price Value Price Value
A S0 6 72
B 7 84 10 80
C 10 80 12 96
D 4 20 30
56 8 64
Solution: Calculation of Fisher's Ideal Index
1980 1990
Commo-| Price Quan Price Quan Piq0 Poqo Piq1 Pogi
diy po tity P1 tiy
41
A 10 6 12 60 50 12 60
B 7 12 10 120 84 80 56
C 10 8 12 96 80 96 80
D 4 6 25 20 30 24
7 56 S6 64 64
357 290 342 284
Time Reversal Test -
Time reversal test is satisfied when
Po X P =1

Epgo ZpAi
2p,4, =357, Zpq, = 342,
Zpdo = 290, pA, = 284.

Po= V357 X 342

290 284

X
ZpA, =284, Ep41 Epao
2pAo = 290
 Statisticy
520
357
Zpg, =342, 2p,40 =
V284 X 290
Pio 342 357
357, 342 X 284 X 290
Po XPn = 290 284 342 357
=V1
Po, XPo=1
Factor Reversal Test i
satisfied when
Factor reversal test is
Ep41
Pa,Po x o1 SpAo
=

= VP4Epg,
P, X

Zpgo =357, Ep,g, = 342

2pA, = 290, 2p4, = 284
342
Po = 35
290 x284
Epi X

24p, = 284, p,g,=342

2poAo = 290, qp, =357
284 342
X
290 357

357 342 284 342

X X X
Po X Qo, = 290 284 290 357
342 342
- 290
X
290
342
290
Ep41
P, xonEpAo
test and
Hence the given data satisfies both the time reversal
factor reversal test.
 Vndex Numbers
521
Illu_tration 23
Calculate Fisher's Ideal Index from the data
given below :
Price
Commodiy Quantity
1989 1990 1989 1990
A 8 10 20 30
B 12 15 10 10
C 6 8 16 20
D 4 6
10

Solution: Construction of Fisher's Ideal Index (M.BA. Rohtak)

Com Price
modiy Quantity

Po q0 41 P190 POqo P1q1 Poq1

A 10 20 30 200 160 300 240
B 12 15 10 10 150 120 150 120
6 16 20 128 96 160 120
D 4 6 10 48 32 60 40
Total S26 408 670 520
Fishers Ideal Index |
Po = }p41 X X 100
}p4% = 526, EpAo
2 p,q, = EpA1
670, 2pA, = 408
ZpA, =520
Po1 V s26 X 670
408 520 X 100
= V1.661 × 100
= 1.288× 100
Po=128.8
Illustration 24
From thefollowing data, prepare, *quantity index numbers" for
the year 1991, taking 1985 as the base year.
(I.C. WA)
 Statistics
522

Comnodiy I Commodity Il Commodiy IIl

Year

Price Quantity Price Quantiy Price Quantity

6 6 3
1985 10

7 7 4
1991 12

Calculation of Index Numbers

Solution :

1985 1991
Commodiry qiPo qoPo q1Pi, qoP1
Po Pi q1

4 12 60 50 48 40
10
7 56 48 49 42
6 7

5 4 24 18 20 15
6 3

Total 140 116 117 97

EqPo X 100
140 X 100 = 120.68
116

QoP) = x 100=117
X100
97

= 120.62

= V120.68 X120.62
= 120.65

Illustration 25
the link relatives given
Construct Chain Index Numbers from
below:

1990 1991
1987 1988 1989
Year :
115 102
100 105 95
Link index :
(B.Com. Hons. Delhi,)
InderNumbers
523.
Calculation for Chain Index
Year Index
Chain Index Numbers
1987 100
100
1988 105 105
X 100=105
100
1989 95 95
105 X 100=90.48
1990 115 115
x 100=121
95
1991 102 102 X 100=88.7
115

Illustration 26

The following table gives the group price index number and
percentage of expenditure on various groups, obtained from inguiries
into the budget of the middle class families of Bombay :
Group %of Expendiure Group Price Index Number
1990, 1992
Food 35 350 440
Fuel and Lighting 10 220 330

Clothing 20 230 400

Rent 15 160 105
Miscellaneous 20 190 340
What changes are seen in the level of cost of living of 1992 as
compared to that of 1990..

Solution:
(BA. Bombay)

Group Price in Rupees Taking 1990 as

1990 %of PxV
1992 P Ependi
Po P base P=x Po
100
Food
350 440 440 35 4399.85
x 100=125.71
350
Fuel and Lighting 220 330 10 1500.00
X 100=150.00

Clothing 230 400 400 20 3478.20

x 100=173,91
230
 Statistics
524
105 X 100=65.62 15 984.30
160 105
Rent
160
20 3578.80
Miscellaneous 190 340 340 X 100=178.94
190
SV= PV =
100 13941.15
£PV
Cost of Living Index =
13941.15
100
= 13941.15
= 139.41

Illustration 2
construction of a certain Cost of Living Index Number the
In the found. Calculate the Cost of
following group index numbers were
Index Number by using (i) the weighted arithmetic mean (ii) the
Living
weighted geometric mean.
Index Number Weighs
Group
350
1. Food
1
200
2. Fuel and Lighting
1
240
3. Clothing
1
4. House Rent 160
2
250
5. Miscellaneous

(B. Com. Bombay Madurai)

Solution :
Arithmetic Mean and Geometric
Computation of Price Index by
Mean
The Consumer Price Index:
By using Arithmetic Mean :
£ IW 2850

PoW = 285
10

By using geometric Mean :

E logI x W
Po =
Index Numbers 525

=
Antilog 24.4016
10
= Antilog (2.4401)
= 15.5
Group Index Number I Weights W IW Log I Log I x W
Food 350 1750 2.5441 12.7205
Fuel and Lighting 200 1 200 2.3010 2.3010
Clothing 240 1 240 2.3802 2.3802
House Rent 160 1 160 2.2041 2.2041
Miscellaneous 250 2 500 2.3979 4.7958
£ W=10 2 W= ELogIx W=
2850 24.4016
Illustration,28
Calculate the index number by the application of Laspeyre's
formula and Paasche's formula from the following data :
Units Consumed Price per unit
Commodity
1986 1988 1986 1988
A 20 16 1.2 2.0
B 35 38 2.1 2.4
C 10 3.0 4.1
D 4 S0 0.8 1.2
And also calculate Fisher's Ideal Index.

Solution: (BA. Bombay)

Computation of Laspeyre's and Paasche's Index
Commo 1986 1988 1988
dity
Po 90 P1
A 1.2 20
Piqo Pogo P1q1 Poq1
2.0 16 40.00 24.00 32.00 19.20
B 2.1 35 2.4 383 84.00 73.50 91.20 79.80
C 3.0 10 4.1 9 41.00 30.00 36,90 27.00
D 0.8 45 1.2 S0 S4.00 36.00 60.00 40.00

Pgo SPogo SPq1 2Poq1

219 163.50 =220.10 =166
 Statistics
326
X 100
Laspeyre's Index (Po) = 2P,40
219 X 100
163.50
= 133.94

Paasche's Index
220.10 x 100
166
= 132.59

Ideal Index = VLaspeyre's Ind x Paasche's Index

Fisher's V+7739.10
= V133.94 X132.59 =
= 133.26

Illustration 29
numbers given below, prepare fixed
From the chain base index
base index number : 1987 1988
1985 1986
1983 1984
103 101
104 98
102
92
the index number for 1984 is
1983 is taken at 100,
If the base at
102. (B.Com. Kerala)

Solution :
Index Number
Computation of Fixed Base
Fired Base
Changed to
Chain Base Chain base Index Numbers
1983 as base
Index
Year Numbers
ndéx
Nunbers
92
1983 92
93.84
92 X 102
1984 102
100
'97.59
1985 104 92 102 X 104
100 100
95.64
98 92 102 104 x 98
1986 100
100 100
98.51
103 X 98
92 102,104, x 103
1987 100 100
100 100
99.495
1988 101 9202 X 104 X 98
100
x 101
100 100 100 100
Index Numbers
527
Illustration :30
The following data relate to the income of the people and the General
Index Number of prices of a certain region.
Calculate: (1) Real income and
(2) Index numbers of real income with 1988.
Year Income (Rs.) General price Index Nos.
1988 800 100
1989 819 105
1990 825 110
1991 876 120
1992 920 125
1993 938 140
1994 924 140
(B.Com, Madurai)
Solution:
Computation of Real Income Index Numbers
Year Income (Rs.)Price Index Number Real Income Real Income Index
Number
1988 800 100 800 × 100 100
= 800
100
1989 819 105 819 × 100 780
= 780 x 100 = 975
105 800
1990 825 110 825 × 100 750
= 750 X 100= 93.75
110 800
1991 876 120 876 ×100 =730
730
X 100 = 91.25
120 800
1992 920 125 920 × 100 736
=736 X 100 = 92
1993
125, 800
938 140 938 × 100 670
=670 x 100 = 83.75
140 800
1994 924 140 924 × 100 660
= 660 X 100 = 82.5
140 800

Illustration:31
Calculate the Laspayre's and the Paasche's Index Numbers from the
following data and comment on the relative merits of the two.
 528 Statistics

Comnodiy Base Year Curent Year

Quantity Price per Kg. Rs. Quantity Price per Kg Rs.

A 10.0 0.80 11.0 0.70

B 8.00 0.85 9.0 0.90

C S.0 1.30, S.5 0.80

(B.Com, Madras)

Solution:
Computation of Laspayre's And Paasche's Indices
Comn Base Year Current Year
odity Po 90 Po1 P190 P191
Price po Quantity|Pricep, uantity
41
A 0.80 10.00 0.70 11.00 8.00 8.80 7.00 7.70
B 0.85 8.00 0.90 9.00 6.80 7.65 7.20 8.10
C 1.30 S.00 0.80 5.50 6.50 7.15 4.00 4.40
21.30 23.60 18.20 20.20

Laspayre's Index X 100

Epolo
18.20
= X 100
21.30
= 85.45

Paasches Index =
~p1 41 X 100
PoP1
20.20
X 100
23.60
= 85.59
Illustration :32
Compute the Cost of Living Index Numbers using both the Aggregate
Expenditure and Family Budget Method from the following information :
Index Numbers 529

Comnodity Unit consumption in Price in base year Rs. Price in curent year Rs.
base year
A 200 10 12
B 30 35
C 50 40
D 20 200 300
40 25 50
F 50 100 150
G 60 20 25
H 40 150 180

(B.Com,, Gujarat)
Solution :
Calculation of Cost of Living Index by Aggregate Expenditure

Commodity Consumpion Price in Rs. Aggregate Expendenure

in base year qo
Base year Po Current Year p1 Poo P190
A 200 10 12 2000 2400
B 50 30 35 1500 1750
C 50 40 S0 2000 2500
D 20 200 300 4000 6000
40 25 50 1000 2000
F 50 100 150 5000 7500
G 60 20 25 1200 1500
H 40 150 180 6000 7200

22, 700 30, 850

Cost of Living Index Epi40 X 100

Epo 40
30,850
X 100
22,700
= 135,903
530 Statistics
Computation of Index Number by Family Budget Method

Price in Rs. Price Weight W PW

Commo
dity Basè Year Po Current Year P
Relatives =Po90
Px100
Po

10 12 120.00 2000 2,40,000

A 200
30 35 116.67 1900 1,75,005
B S0
40 50 125.00 2000 2,50,000
C S0
200 300 150.00 4000 6,00,000
D 20
25 S0 200.00 1000 2,00,000
E 40
100 150 150.00 S000 7,50,000
F 50
20 25 125.00 1200 1,50,000
G 60

H 40 150 180 120.00 6000 7,20,000

EW= SPW =
22,700 30,85,005

2PW
Cost of Living Index X 100

30,85,005 X 100
22,700
= 135.903
Illustration :33
indeX numbers
Prepare fixed base index numbers from the chain base
given below:
Solution :
Construction of Fixed Base Index
Fixed Base Index
Chain Base
Year
94
94
1989
104 × 94
104 =97.76
1990 100
104 X 97.76 101.67
1991 104
100
93 X 101.67
1992 93 = 94.55
100
103 × 94.55 =97.39
1993 103
100
102 × 97.39
1994 102 = 99.34
100
IndexNumbers
531
Year 1989 1990 1991 1992 1993
1994
Chain Index 94 104 104 93 103
102

Illustration:34
(B.Com., Kerala)
The following are the prices of
Calculate a price indexxbased on price commcdities in 1990 and 1995.
relatives, using the geometric mean:
Year Commodity
A B C D E F
1990 45 60 20 50 85 120
1995 55 70 30 75 90 130

(B.Com, Bharathiar)
Solution :
Computation of Price Index
Commodity 1990 Po 1995 P, P 100 P LogP
Po t
A 45 122.22 2.0872
B 60 70 116.67 2.0669
20 30 150.00 2.1761
D 50 75 150.00 2.1761
85 90 105.88 2.0248
F 120 130 108.33 2.0348

£ log P =
12.5659

Po1 |= Antilog (Elog

N
P

= Antilog 12.5659
6
= Antilog 2.0948
= 124,4
Illustration :35
The cost of living index number on a certain date was 200. From the
Dase period, the percentage increase in prices were :
Rent-60,
lothing250, Fuel and Light-150 and Miscellaneous 120. What was the
percentage increase in the food group ?
The weights for different
groups were :
Food-60; Rent- 16; Fuel and light-8: Miscellaneous 4, Clothing-12
 532 Statistics
Solution:
ComputationsofIndex
Curent lndex I Weight V IV
% increase in
Group
Price
X 100 + X 60 60 (100+x) =
Food 6000 + 60x

160 16 2560
Rent 60
350 12 4200
250
Clothing
250 2000
Fuel and Light 150
320 4 880
Miscellaneous 120
2V= 100 2IV = 60 X
+ 15640

or
Cost of Living Index of Current Year = 200
EIV
= 200

60× +15640 = 200

100
60 x +15640 = 20,000
60 X = 20,000 15640
4360
X=
60
X = 72.67
72.67
The percentage increase in food group was
Illustration :36
and cost of living index
The following table gives the money wages
numbers based on 1989:
Year 1989 1990 1991 1992 1993 1994 1995
65 70 75 80 90 100 120
Wages (Rs.)
Cost of living
250
Index Nos. 100 110 120 130 150 200
Calculate the real wage index numbers.
(B. Com. Madurai)
Solution :
using the following
Real wages for different years can be obtained
formula.

Real Wage Money Wage X 100

Cost of Living Index
1989 as base.
Real Wage Index Numbers are calculated by taking
Index Numbers
533

Calculation of Real Wage Index Numbers

Year Money Wages Rs. Cost of Living Real Wages Rs. Real Wages Index
Index
Numbers.
1989 65 100 65
X 100 = 65 100
100

1990 70 110 70
x 100 = 63.64 63.64
110 x 100 = 97.91
65
1991 75 120 75 x100 = 62.50 62.50
120 x 100 = 96.15
65
1992 80 130 80 61.54
X 100 = 61.54 X 100 = 94.68
130 6.5
1993 90 150 90 60
X 100 = 60.00 KE X100 =92.31
150

1994 100 200 100 50

x 100 = 50.00 x 100 = 76.92
200 65

1995 120 250 120 48

X 100= 48.00 x 100 =73.85
250 65

Illustration:37
Itis stated that the Marshall Edgeworth Indexis agood approximation
to the ideal index numbers. Verify using the following data :
Commodity 1994 1996
Price Quantity Price Quantity
A 2 74 3 82
B 7 125 4 140
7 40 6 33
(B. Com, Bombay)
Solution:
Construction of Marshall Edgeworth and ideal Index Numbers.
Commodity P P1 41 Po 90 Po91 P190 P191
A 2 74 3 82 148 164 222 246
B 125 4 140 625 700 S00 560
7 40 6 33 280 231 240 198
Pofo= 1053 p10 Zp191
= 1095 =962 = 1004

Marshal Edgeworth Index = X 100

Epo 90 +2po 91
534 Statistics
962 + 1004 1966
X 100 = X 100
1053 + 1095 2148
91.53

Fisher's ldeal Index X 100

Po41

962 1004
X 100
1053
XT095
965848
1153035 X 100 = V0.837657 x 100
= 0.91523 × 100
= 91.523
Marshal Edgeworth Index is a good approximation to the Fisher's
Ideal Index number.

Illustration 38
From the fixed base index number given below, prepare chain
base index numbers :
1991 1992 1993 1994 1995 1996
94 998 102 95 98 100

(B. Com. Madurai)

Solution :
Computation of Chain Base Index
Year Fixed Base Link Relative Chain Base Index Number
Index Number

1991 94 94 94

1992 98 98 X 100 = 104.25 94 X 104.24

= 97.99
94 100

1993 102 102

X 100 = 104.08
98 X 104.08
101.99
98 100

1994 95 95 102 x 93.13

102
× 100 = 93.13
100
94.99.
1995 98 98 95x 103.15 = 97.99
X 100= 103.15
95 100

1996. 100 100 X 100 = 98 × 102.0499,99

102.04
98 100
IuderNnhers
535
FORMULAE
A. Unweighted Index
(a) Simple Aggregative Index
ZP,
PoSP, X 100

(b) () Simple Average of price Relative Index

EP+2 Px 100
Po,= N
(ii) When Geometric mean is used :
E logP
Po = Antilog N
B. Weighted Index Numbers
(a) Weighted Aggregative Index:
(i) Laspeyre's Index or PoLa) = X 100

(i) Paasche's IndeX : or Pos(pa) = SPA1 100

SPA
(i) Bowley's Index:
EP4%. ZP4
2Pe,go ZPa
Po(8) 2
L+P

(iv) Fisher's Ideal Index

X X 100 or
Poe) = 2PA2PA
= VL XPX100
(V) Marshal Edgeworth's Index :
ZP{q, +4)X 100
Pol. (Ma) =
(vi) Kelly's Index :
Por(K) x100 whereQ=
2PA
C. Weighted Average of Price Rclative Method
2 PV
Po =
D. Consumer Price Index :
 536 Statistics
()Aggregate Expenditure Method
Consumer Price Index EPo X 100
2PAo
(ii) Family Budget Method :
Consumer Price Index =
2PW
2W
EIW
(ii) Consumer Price Index = £W
E. Time Reversal Test :
Time Reversal Test is satisfied when
Po XPo=1 or
XP41 X
PA1 X
£P, =1
2PA EPA
F. Factor Reversal Test
EP4 2P40
Factor Reversal Test is satisfied when

Po X lo1 = 2P41
2PA0

THEORET0CAL QUESTIONS
1.
"Index numbers are economic barometers". Explain the statement and explain
what precautions should be taken in making use of published index number.
(B. Com. Madras).
2.
What is an Index number ? Why are index numbers called "Economic
barometers" ?
(B. Com. Bombay,)
3. Analyse the problems in the construction of index numbers and comment on the
need for weighting.
4
(B. Com. Madras)
What are the tests of a good index number ? Define Fisher Ideal index number
and show that it satisfies all these tests.

5.
(B. Com. Madurai)
Discuss briefly the uses and limitations of index numbers of
prices.
6 (B. Con. Bombay )
Distinguish between fixed-based andchain
points of importance in choosing the base inbased index numbers. What are the
the determination of cost of living
index numbers ?

7.
Explain how a cost of living index number is (B. Com. Madras)
formula used to construct such an index constructed and examine critically the
serve? number. What purpose do such indice
(B.A. Econ. Bombay)
nlexNUmbers
537
required to construct a cost of living
You are What
8. Bombay. information will you collect for theindex for the textile workers 1:
index ? purpose ? Explain the methoo
of constructing the
pefine () Laspeyres' (ii) Paasche's and (ii) Fisher index(BA.number
Econ. Bombay)
of
Comment on the statement, "LaspeyTe OVer prices.
under estimates them in estimates the price changes while
Paasche general, and hence Fisher provides a better
actimate than both Laspeyre's and Paasche's"
(B.A, Econ.Hons. Calcutta)
PRACTICAL PROBLEMS
Calculate current year price index number by using various formulae from the
10.
following
Base Year data Current year data
Commodity Price per unit Quantity Price per unit Quantiy
1 10 15
A
B 5 12 6 10
10

[Ans. PoiLa) = 124.54, PoiPa) = 124.32 Pa(F) = 124.43)

11. Calculate Fisher's Index Number for the data given below :

Base year Base year Current year Current year

Commodity Oly in (Mds) Price in (Rs.) 9ty in (Mds)
Price in (Rs.)
A 50 10 S6

2 100 120

4 60 6 60
C

12 24
D 10 30
See whether this satisfies factor reversal test and time reversal test.
(B. Com. Annamalai)
(Ans. Po1 = 136.8)
numbers
12. Calculate (a) Paasche's index, (b) Laspeyre's index and (c) Fisher's index
for the following data :

P q1
Commodity Po 40
15 25
A 12 20
10
B 16
10
1
2 12
15
1
D 1 65
60
1
10
3 2
(B.Com Madras)
129.60)
(Ans. (a) 130.13 (b) 129.09 (c)
 538 Statistics
13. From thc tollowing data calculate price index numbers for 1990 with 1981 ab vaNs
by (i) Laspcyre's method, (i) Paasche's method, (ii) Marshall-Edgeworth method
and (iv) Fisber's method :
1981 1990
Comnodiry
PO P1 qi
A 20 40 6

B 10 60

C 40 15 S0 15

D 20 209 20 25

(B.Com. Delhi)
(Ans. () 124.70, (ii) 121.77, (ii) 123.3 (iv) 123.2)
14. The following data relate to the prices and quantities of five commodities in the
years 1989 and 1990. Construct the following index numbers of price for the year
1990 by taking 1989 as the base year. (a) Laspeyre 's index (b) Paasche's index, (c)
Fisher's index, (d) Marshall's index, (e) Bowley's index and Kelly's index:
1989 1990

Comnodity Price Quantity Price Quantity

15 125 13 150
A

100 15 80
B 12

150 10 125
C

200 250
D 10
80 60
6

100.48, (e) 100.48 () 100.43)

(Ans. (a) 102.81 (b) 98.16, (c) 100.46, (d) 100.43, (e)
of six commodities, Construct
15. The following data relate to the prices ad quantities
the following indices :
Fisher's index (d) Bowiey's index (e)
(a) Laspeyre's index, (b) Paasche's index, (c)
Marshall's index () Kely's index.

P q1
Commodity Po q0

3 18
14

6 25
2 8 18
40
3 3 25

12 48
15 36

7 18
14

5 19
6 7 13
(B. Com. Madurai)
Index Numbers
539
(Ans. (a) 73.61 (b) 73.05 (c) 73.22 (d) T3.19 (ey 13.29 y 96.96)
16. From the following information, construct (a) Laspeyre's index, (b) Paasche's
index, (c) Fisher's index, (d) Bowiey's index, (e) Marshall's index and () Kelley's
index.

Commodity Po q0 P1 q0
Rs. Rs.

1 15 14 18 10
2 16 18 19 15

3 19 35 25 20
4 24 39 29 30
21 40 25 35
6 16 31 18 25
(Ans. (a) 121.04 (b), 120.38 (c) 120.70 (d) 120.71 (e) 120.75 () 120.75)
17. Construct the Cost of Living Index Number of 1990 using the Family Budged
Method.

Expenses Base Year 1986 Current Year 1990

Food 40 150 174
Rent 15 S0 60

Clothing 15 100 125

Fuel 10 20 25
Miscellaneous 20 60 90

(Madurai, Kamraj)
(Ans. 125.65)
18. The following table gives the date of
production in tons andprice per ton of 4
principal crops in India, during the year 1980-81 and 1989-90.
base, construct Fisher's Ideal Number of price for Taking 1980-81 as
l989-90.

Production (Lakhs of Tons Price) (Rupees per Ton)

Commodiy 1980-81 1990-91 1980-81 1990-91
A 250 300 150 130
B 100 120 120 200

20 30 600 1000
D 10 22 200 300
 Statistics
540
(BA,, Econ. Madurai,)
(Ans. Po1=120.3)
in 1985 and 1990. Calculate a price index
19. The following are prices of commodities
geometric mean.
based on price relatives, using the

Commodity
Year C D
A

20 50
46 50
1985
30 75
55 70
1990
(B.Com. Bombay,)
(Ans. 124.2)

lIdeal Formula and show that it satisties

20. Compute Index Number using Fisher's
Reversal Test.
Time Reversal Test and Factor

1988
1987
Comnodiny
Price Quantiy Price
Quantiy
15 12
12 10
A
15 7 20
B
20 9
C 24 5
14
D 5 16
(M.Com. Nagpur, Rohtak)
(Ans. 11S.76)
following data and comment on it :.
21. Calculate Fisher's Quantity Index from the
Current Year
Base Year
Price (Rs.) Value
Price (Rs.) Value
Commodiy
12 300
A 10 200
10 220
B 8 108
25 250
C 20 160
20 140
D 18 144
30 300
35 280
(B.Com. Bombay)
(Ans. 103.88)
Index Numbers
54]

22. From chain base index numbers given below, prepare fixed base index numbers :

Year 1981 1982 1983 1984 1985

Index Numbers 80 110 120 105 95

(B.Com Kerala)
(Ans. 80 ; 88; 105.6; 110.9; 105.3)

23. From the data given below construct an index number (i) with 1980 as base (ii)by
chain base method :
Price of commodity from 1980 to 1985
1980 1981 1982 1983 1984 1985
50 60 62 65 70 78
(Ans. (i) 100; 120 ; 124 ; 130; 140 ; 156.
(it)( 100;120; 124; 130; 140; 156.
(Answers are the same because the indices have been computed for one
commodity only).
24. What are the uses of the cost of iving index number ? Calculate the cost of living
index number from the following data :

Price
Items
Weight
Base Year Current Year
Food 30 47 4
Fuel 12 1

Clothing 14 18 3
House Rent 22 15 3
Miscellaneous 25 30 1

(B.Com. Madras,)
25. In the
(Ans. 123.91)
construction of a certain cost of living number the following group index
numbers were found. Calculate the cost of living index number by
weighted arithmetic mean and (ii) the weighted geometric mean. using () the
Group Index Number Weights
1. Food 350
2. Fuel and lighting 200 1

3. Clothing 240 1
4. House rent 160 1

5. Miscellaneous 250 2
 542 tatistics
(S. Com. Bombay,)
(Ans. (1) Po1= 285, (2) Poj= 275.4)
26. From the chain base index numbers given below prepare tixed base index numbes
and verify the answer :

1981 1982 1983 1984 1985

Year
110 166 140 200 150"
Index
(B. Com. Delhi,)
(Ans. 110, 182.6, 255.64, 511.28, 766.92)
27. The following are the group index numbers and the group weights of an average
working class family's budget. Construct the cost of living index number:

Groups Food Fuel and lighting Clothing RentMiscellane.

oUS
Index Number 352 220 230 160 190

48 10 8 12 15
Weight
(LC. WA.)
(Ans. 276.4)
28. The price quotations of 5 ifferent commoities for the years 198S and 1990 are
as follows :

Price in Rs.
Commodity Weight 1985 1990
!

A 4 2.50 4.00

B 8 2.80 3.20

C 6 3.20 4.00

D 2 3.00 4.50

1.00 1.70
Calculate the price index number for the year 1990 with 1985 as the base year,
using :
) Simple average of price relative, and
(B. Com.
(ii) Weighted average gf price relatives.
Poone,)
(Ans. (1) 153, (2) 160.8)
29. Compute Chain Index Numbers from the ink relatives given below :
Year 1988 1989 1990 1991 1992

Link Index 100 105 95 115 102

(B. Com. Delhi,)
(Ans. = 100, 99.7S, 114.71, 117).
 S43
Index Numbers
Index
30. From the following fixed base index numbers, calculate Chain Base
numbers :
1985 1986 1987 1988 1989
Year
Fixed base Index Numbers 188 196 204 190 196

Chain base Index numbers = 100, 104.3, 104.1, 93.1, 103.2)

From the following Chain base Index numbers (from year 1985 to 1989) calculate
31.
fixed base index numbers (taking index for 1985 on fxed base as 188.) :
100, 104.3, 104.1, 93.1, 103.2.
Fixed base Index numbers = 188, 196, 204, 190, 196)
32.
Construct Index Number for cach year from the following average price of cotton
with 1979 as the base.
Year Price (Rs.)
1979 75
1980 50
1981 65
1982 60
1983 72
1984 70
1985 69
1986 75
1987 84
1988 80
(B.Com.Jodhpur, Madurai)
(Ans. 100; 66.67; 86.67; 80; 96;93.33 ; 92; 100 ; 112; 106.67)
33. Construct a price index for the following by (a) Simple Aggregative method, (b)
Average of price relatives method by using both arithmetic mean and geomeiric
mean :

Commodity: A B C D E F

Price (1975) Rs. 20 30 10 25 40 50

Price (1985) Rs. 25 30 15 35 45 55

(B. Com Madurai, Calicut)

(Ans. (a) 117.14; (b) 122.92 and 121.7)
34. Calculate weighted index number for 1981 from the following :
Item Cuantity Price (1980) Rs. Price (1981) Rs.
A 20 units 200 320

B 14 units 400 420

C 1S units 100 120

D 18 units 40 60

E 10 units 20 28

(BCom Bangalore, Bombay)

(Ans. Po1 =137.27)
 344 Statistics
35.
Construct index for 1988 taking 1985 as base from the following data by using ta1
Arithmetic mean (b) Geomnetric mean and (c) Median.

Commodities A B C D E F

25 50 8.62 24.6 1s
Price in 1985 (Rs.) 40

Price in 1988 (Rs ) 60 31.25 375 8.62 18.45 11.25

(B.Com. Madurai)
(Ans. (a) 100 ; (b) 97.1; (c) 87.5)