r-gc -.

Nr2oi2
F-DTN-M-TUBB

STATISTICS Paper II
Time Allowed : Three Hours Maximum Marks : 300

INSTRUCTIONS
Each question is printed both in Hindi and in English. Answers must be written in the medium specified in the Admission Certificate issued to you, which must be staled clearly on the cover of the answer-book in the space provided for the purpose. No marks will be given for the answers written in a medium other than that specified in the Admission Certificate. Candidates should attempt Questions no. 1 and 5 which are compulsory, and any three of the remaining questions selecting at least one question from each Section. The number of marks carried by each question is indicated at the end of the question. Assume suitable data if considered necessary and indicate the same clearly. Symbols I Notations used carry usual meaning, unless otherwise indicated. Charts I Figures, wherever required, are to be drawn in the answer book itself and not on separate graph sheet.
IMPORTANT : Whenever a question is being attempted, all its parts/sub-parts must be attempted contiguously. This means that before moving on to the next question to be attempted, candidates must finish attempting all parts/sub-parts of the previous question attempted. This is to be strictly followed. Pages left blank in the answer-book are to be clearly struck out in ink. Any answers that follow pages left blank may not be given credit.

Ezi7W

31 f ae" a5r feed) .041-cK '1`111Y-f-rff 041 t

RT.O FEU tg

SECTION A 1. (a) Describe single and double sampling plans used in acceptance sampling. Define the operating characteristic function ola sampling plan. (b) The cost of maintenance of a machine is given as an increasing function of time and its scrap value is constant. If time is measured continuously, then show that the average cost will be minimized by replacing the machine when the average cost to date becomes equal to the current maintenance cost. (c) Describe how the control chart based on sample ranges is constructed. Describe Type I and Type II censoring. Discuss the consequence of such censoring in the analysis of lifetime data. (e) Show that in a linear programming problem, the feasible solutions form a convex set.
5x12=60

(d)

2. (a) A person needs at least 10, 12 and 12 units of chemicals A, B and C respectively for his garden. A liquid product available in the market contains 5, 2 and 1 units of A, B and C respectively per jar. A dry product contains 1, 2 and 4 units of A, B and C per carton. The price of the liquid product is 3 per jar and that of the dry product is 2 per carton. How many jars and cartons should be purchased to meet the requirements and to minimize the cost ? F-DTN-M-TUBE 2

30

[Contd.]

ti c1I a1M ar .crw@ 1,tur-AFdtityr 1. () tv p- t'W I ts 4[311 r qui for iArcitRI-f 374'7 d7 1c1 k h 3ifi47ual th , 9HT A cititH 37-41
1iFolfrEm

irth q4-77 12FF.9-

,(@-Tura

wrzr

1,

to-r # 3fr{ Tkr* 7 tik wrzt :17f 3T97 (kr) 1 AflTWTtM TirrTRi to- diim Trt9- T[ aifrlfl c114Ici ,11 714 3T1 c14, Tr4R wr 6'1 4 1/2f 9-11.-14,101, JI-PT4cr 1/2 4k14( '6)cl? 441 -0r (71 ufrazl tr-{ aimiRcr f d-77 f1/2 Af (TO qui7 tr tdfr1f 3/44 -i# (a) 5[2 firRitiur # *47-1/27 , TRI ifttrifit tr auf -nrF
-q)-72:F

rq qu'

q,ch TrETIN -11-kctitt Qldr 2. (1/2) t;4-) C 1/21

5x12=60

1/21 31I qiil;:f 1/2 fyikk

1iii441:147i A, B

4h4-1--#-4h4-1 10, 12 AT 12 TETTA

I 41s311t Ti Tur kon dkcer 3114711/2M 611 5, 2 C 1/2 cf 7T{ A, B 3 ref 1.41tiT2r iIT Afd. 4-Ai.4 A, B 3Tt{ C 1 T NEfi c I Tk* 11/2 t. I (Ikci 4 r f 1/21 Afd . 4 #T1/2 9h+1YI: 1, 2 3417

t2 mrd 3 't aih -TEF -tic1H r4 3tR c1I'Id dr3if Ei trTr T%3TraTT*m4,4 1/2 RV f1/29-4 1111 3-th Th19. 4.14.1c4 aIF L& ? :7 , YIlt 1/21 1
F-DTN-M-TUBB 3 EContd I

(b) Consider a zero-sum game between two persons A and B, where each player has three strategies At , A2, A3 and B1, B2, B3. Player A's payoff matrix is given below : Player B Bt Al Player A A2 A3 15 6 7 B2 2 5 4 133 3 7 0

Is there a saddle point 7 What are the optimal strategies for the two players ? 15

(c) Arrivals at a telephone booth are considered to be according to a Poisson process with an average time of 10 minutes between two consecutive arrivals. The length of a phone call is assumed to be exponentially distributed with a mean of 3 minutes. What is the probability that a person coming to the booth will have to wait ? What is the average length of the queues that form from time to time ? 15

F-DTN-M-TUBB

[Contd.]

.64fTht A 3t B

4-C-41 4-a. : krT .?1

-1 a did TM711A, , A2, 4111, 41e 51 -T P.Ac415 c4114 , 132, B3 e I figc-n,slAw ISCd7A3 ath Bi 4114 f -zrr 'Rif t : ft B1 Al -a1 1 A tf A2
A3

131 B B2 2 5 4 13 3 7 0 ? ? 15

15 6 7 fq3 ci-411

r4'41

Lm

- 7 TT 7 373* (11) 1,.:b 1 -d 1Th - 1 io friTe c;f1 lzfr14 &TR aiTTu chic4 TTRT 1 -7 1;ch Th mik1-11c114;itM:

*Tt 3T1Trff 3 fk9Z

,-11c-11 t INIlr-NA-If c14-11

g
.? rtn

7 3174 qrA 1RTRI-iA114 tit (4-1 ar

@-1711t ifrrm -4t

ofw
15

?
F-DTN-M-TUBB 5

(Contd.'

3. (a) A sample of 100 screws was selected on each of 25 successive days in a factory manufacturing screws, and each screw was examined for defects. The data for number of unacceptable screws on different days are given below : No. of I defective Day screws 7 4 3 6 4 9 6 7 5 3 7 8 4 14 15 16 17 18 19 20 21 22 23 24 25 No. of defective screws 6 2 9 7 6 7 11 6 7 4 8 6

Day . . _ 1 2 3 4 5 6 7 8 9 10 11 12 13

Assume that the production process was in control during the period. Determine the upper and the lower control limits (UCL, and LCL) based on this data for the proportion of defective items.
30

F-DTN-M-TUBB

[Contd.'

3. () atit chi

kciZt a 25 3rit [1( 4- a r -ciWe 141c1c0 kr9- too reifeu iiru 4 1 t11 I c) ' cur, &it{31#g f Rrit ut awft- rzi ij4 r thstii

ft; 41
_ .

fi-r 1 2 3 4 5 6 7 8 9 10 11 12 13

TrOzf k -carr 7 4 3 6 4 9 6 7 5 3 7 8 4

froi 14 15 16 17 18 19 20 21 22 23 24 , 25

'FFq)-Er ert .942/4 6 2 9 7 6 7 11 6 7 4 8 6

chc.-Hir
feu N 2 3 3-77. fyit airR 7z7r

sw:r arofq
37N-CR9 CrEm J-k)f TIT Ti#11 (UCL f4430

P7I LCL) M-9frm- tr

[Contd.!

(b) Consider an acceptance sampling plan where 50 items sampled from a huge lot will be examined and the lot will be accepted it at most two of the sampled items are found defective otherwise the lot will be rejected. Evaluate the acceptance probability as a function of the proportion of defectives in the lot and sketch the Operating Characteristic (0C) curve. (Only a rough sketch is required and no graph paper is necessary) 30

4. (a) The following table gives the maintenance cost per year and the resale price in of a certain machine whose purchase price is 5,000 : Maintenance cost 1500 1600 1800 2100 2500 2900

Year 1 2 3 4 5 6

Resale price 3500 2500 1700 1200 800 500

In which year is the replacement due ? F-DTN-M-TUBB 8

30 [Contd.]

(-W

. Tin Ml[ R fW{44 -) ta1a,tur Afciag.i 3117)771 ir ff 50 0k1 wr[4. 3144 d4 aft 9414ui # 3ift4T7T4 3fr( Aqai ?1 44 3-fl' r4 4r4 AlaP, 3TR2u A, IfE0 -cra't 1411 I ttl1aht'+r 3fli44bor mrr sKat T1sr \-t d4 # P faisbirci at th 3944 a-a. -a4I (% 311111Ti aph 4ff )-{ .7t{aicndot .4{ la =IWO, 31) t) 1-12 Atntd Tean(
30

-rm z 5,000 t, srth 1 --q T1 17117 7, fi:rt u4 4. (7) f7Tft . QTrn pg-tura T1 oind 347 (z fai Trruift # m-mr-4 MI'lci 1500 1600 1800 2100 2500 2900 7 :a:r 71 d3500 2500 1700 1200 800 500

44 1 2 3 4 5 6

fir F-DTN-M-TUBB

# aka e ? 9

30

[Contd.]

(b) Consider a Markov chain with three states 0, 1 and 2. The state 0 is an absorbing state, while from any of the other two states the chain moves into any of the three states with equal probability (i.e. p(a I b) = 1 for a = 0, 1, 2 and b = 1, 2). 3 Starting from either state 1 or 2, let N = the number of steps or transitions needed before absorption into state 0 takes place. Derive the probability distribution of N. What is the expected number of steps or transitions before absorption into state 0 ?
30

F-DTN-M-TUBB

10

[Contd.]

(E

0, 1 3th 2
4-114

3raP,r3

1N 41TIFT uT2T . 4cf) 1:1*y

ci1Riu,

I mai2fr

o l&cf-} 31-TA* 3TqT2IT * -t 4 rift sipA7r 31-q7.9-ini)- a-4 f(31.211N, p(a I b)

)414i44-113-ff 3N\T b = 1, 2

1 a = 0,1.2 3

1). # f(4110 c12411 31W-21-13-

4 f-

4) 4 717y4T

5ir

C 7iT Rra t4 1 43-1.7-NT 1 NT 2 f

-{ j n

4-114-11

N SINT-NT 0 i=t 3-14* NT 4--Traff 1;1 k-i{9q! 1 9 0:1N TIT 4oTu-ff

6)4 4

3Tra-Tzg trt N cid4i 14

.rN7 1 31NT-NT 0 .4 4t9.4i

F-DTN-M-TUBB

11

[Contd.1

SECTION B 5. (a) For the AR(1) process defined as X i = 0.6 Xt _ + E t, where t = I, 2, 3, ..., X0 = 0 and F1 , 2, Ea , ... are independent N(0, a2 ) variables, what is the expression for the spectrum (i.e., spectral density) ? (b) Describe the problem of multi-collinearity in multiple linear regression. Flow does it affect parameter estimation ? Describe some solution for this problem. (c) Describe Laspeyres' and Paasche's methods (hr computing price indices. Discuss how Marshall Edgeworth formula leads to a compromise between the two methods. (d) Describe the exponential growth curve. Explain how one checks whether a given time series has exponential growth pattern and how one can fit an exponential trend to a time series. (e) What is the percentile score of a student in a test ? Suppose that the original scores of the student in a test are approximately normally distributed. Then what will be the mean and the variance of the percentile scores ? 5x12=60

F-DTN-M-TUBB

12

[Contd.]

5. () AR(1) TOT

m fll& X t = 0.6 X t _ + e t, A E1, E9, 63,

1-4Thrrii7
3-1'6.1 t NW, 1, 2, 3, ..., =0

62 )
?

A47:1 ;

(45)

wghi m1 4:1441 quh 7TcrT 'tPT cig+ -01 737 UT T3TirqU Trfft Cd-R. tfA-7 I ,1141c4 31T? eo a al&

tfw.r,
(Tr) itc-4 ttichicht ' fqrgzff trrit Nr9W279. titlf 1r31Q, rift; 3fT

gen. trf77 421 4 11 fqfgzi't 44RicerT7ti 1ft

'174

Wita

(E) <-14.q1clich

47 I c=z4IC9/-1[ -9. 4tF MI 111

1 -'4 1.;*

W-F
4-} TR (4,

=la-4 qtr fcit4

4;4 3-T1#97

. *-ichcit ! iI

La tRtiWr
eddr
?

onceLvir

fau42ff trwh 1-14%/7 ails

r .1c1c14-1(sh TITiTh

it*

ItaTiTr

fair
I ci
5x12=60

To
11c.1c141011

chd d : N*-114-11-tIci:

4F -a-ff .*

34444ur a-44r .61't ?

F-DTN-M-TUBB

13

[Contd.]

6.

(a) Consider the linear model yLi =ui +(li +Eii,i= 1, ..., I j = 1, ..., f1- j's are independent N(0, a 2) variables. Obtain the least squares estimates of a, as and fip 0,1 lbr r, s = 1, ..., I, r s and p, q = 1, . ,J,pxq. Obtain also the variance expressions for those least squares estimates. (b) Derive the normal cquations for fitting a linear trend to time series data. Suppose that a time series Xt has been observed at time points t = + 1, 3, + 5, + 19, 21 and for this series and
t

30

xt =
t

528.2

t X t = 1956.4.

Determine the linear trend equation that will fit the data. 30

7.

(a) What is the Logistic curve ? Describe in detail the method of Pearl and Reed and also Rhodes' method for fitting a logistic curve to population data. (b) Define Irving Fisher's "ideal" price index. Explain why it is called an "ideal" index by giving relevant mathematical details. Give an example of well-known price-index formula that is not "ideal" and explain why it is so. 30 [Contd.) 30

F-DTN-M-TUBB

14

6.

() TIR M1fmt triffT

yo = cei +
Eu's talc-14 N(0, a2 )I ar a s 3117 ISp - ISo r, s = 1, ..., I, r s 311t p, (I = 1, p

-g7 13 -1'1cia of 3TlTf sr7r tg

cirsw, f* i c *Id ;Pin xt *E@ fq3A t = + 1, 3, + 5, ..., 19, 21 7. Od kit; x t = 528.2 3-17 5 ;1.aff
mu'. t X t = - 1956.4 1 - f77 ffiNircu *

fr c-R
ath '

-*

d 31RM. f14I4hkuf 30

A7 jiai I qa N A4f

-ef 4 c* ? ,-N-R-R9qr A* 41 :Th. eic* *Err slot 7. () 11 4c' mil as fariT AU44 If d-crm a 30 I A{ Ns -:{:f fafq J quh fatiR -TiR (g) TM-Tr fch
"3-1WRI Hj.-4 ttliznion Z Ifoiffird1 -f A7 iiFuk-114 faaroT k7 4:7 * *sok-1r .3qt{ui I la, 30

rr -A7 - I kr* "3{{q71" T-Rzr "3{d" -let* 3fr{ F-DTN-M-TUBB

kfrr aqt
15

TYA7

'Contd.]

8.

(a) Give a brief outline of factor analysis and discuss its importance m Psychometric studies.

(b) Distinguish between an abridged life table and a complete life table. Discuss a method of constructing abridged life table.

cc) What is reliability of test scores and how is it determined ?

(d) Write a note on official agencies responsible for data collection on trade and prices. 4x15=60

F-DTN-M-TUBB

16

8. () 3467PA m 7

Tckur Tck tif f

#FP,97r ffifg

4-tiltult 3117

d-1 1 'q q4-qpsth If34-

I Sr atr-wit 4;1 ktiHr

fr

(1r) trtivo-r W4i fRlat-d- 61(11

foRcRidtior WEff ?

afR

(Er) 041141t 3fr{

4zr m c1 tr{Gd TiVZ l4, faarttrk 1

Fl
4x15=60

F-DTN-M-TUBB

17

F-DTN-M-TUBE

i col FF13-1:11' II
TrlTer cf EfUt
arlaVE w-t 11. 4134T 3ti1t
guTM

3(1/1

3t TITEZITT 4 0-0 n,rcf fa F&T an-qt ridw-w TY WC( ileil et 37k s* Tfitarr mi -titc sth-g-RTI&T & To. 59z5- EIT 3a -if Wit- TP-179" rff
cr331 3nC

WeTI ai, 1, ra' I TMW-U ER- icrtRact 47aPT W?:& TilerPT Wirt T 31W -157 ffl4fr I W-1 \LIS,' 1 31)7 5 3IP arcit r a77fi7;774 4 cite c154-#-(1)4ew viler gr4th ariilf I Auldr1q. t'&1 7 ft(17,31374achn Fla 377draw e 3crejw say &e &T visa er Ef ffe7T Pffr5z- & 7 - 0u TroVm.c),(1,11 * gedff o1 Q5 3R:reff 3-F-T f ell
e /1=1, Jig 377dt1TIM" :Bch t)q< 97 3d - i.
m dn2 r crz. rq7$L 7,

pr e

377eltrn $itA* rivR. a5r drt< aV Tif Arff WW1 $1777Y/39417711 i itrn W127-Wf27 rt1W 3121 zit " g 31714 TRW ai7 driC AZ44 ifU 3711 00 g l Pod ItqW VIT4/317477771 WWI 011(7 I Ti7Y5 riwvar &nu I driC midi wen tie u yr& gzigu dm) 31 e RE ong, 21epr77 gI

ma rgui

37ev1

Note :

English version of the Instructions is printed on the front cover of this question paper.