CHAPTER ONE

1.0 INTRODUCTION

1.1 BACKGROUND OF THE STUDY

In the last census exercise conducted by the National Population Commission

in 2006, Nigeria’s population was put at nearly one hundred and forty million.

This is undoubtedly an alarming figure for a country with distressed

economy. The consequences of overpopulation are well known, with

characteristics of Socio-economic problems such as unemployment or

underemployment, low level of per capital income, low standard of living,

poverty, huge external debt burden and many more social vices becomes

manifested.

An important area where data collection is essential is human population. Its

data collection procedure is usually by direct enumeration of every individual

in the territory at approximately the same time. Collecting data on human

population involves a complex series of related activities. There is no

gainsaying the fact that accurate population statistics is vital to sound

development planning and economic management. Apart from information on

1
the stock of the country’s population, it is essential to know the rate at which

the population is changing, structurally and in the aggregate.

Provision of good health services and facilities has been advanced as a major

cause of high birth rate. Adequate and affordable health services enhance the

chances of safe delivery, reduce miscarriages, still births and also reduce

infant mortality.

According to Jamison (2018), improvement of life expectancy especially in

developing countries like Nigeria and the sharp drop in mortality can be

attributed to the huge investment in the health sector by most government

especially in the developing countries. His opinion reflected in the case of

South west, considering the huge resources already pumped into the health

sector by the government. It is pertinent to reflect in this study that, health

care services for infants up till 12 years in most South west state is now free

without any encumbrances. He stressed that improved health increases the

enrolment of children in school and make them learn better and generate for

the society able bodied and skilled man-power needed for promoting

economic growth.

2
African sociologists have a different view concerning the reason for high birth

rates. According to them, some religious groups such as Islamic religion

permit a man to marry more than one wife and which is often a precondition

for the accord of certain rights and privileges. In the cultural explanation,

certain rights and honours are onlyconferred based on the number of wives

and children a man has. According to Otite and Ogionwo (2015), polygamy

(marriage of one man tomore than one woman) is a common indigenous

system of marriage in most if not all African countries. According tothem,

some traditional rulers for example in Cameroon were said to have had as

many as 300 wives. Theymaintained that polygamy is practiced in almost all

the religions in Nigeria.

Further causes of high birth rates are high fertility and low mortality.

According to Fapohunda (2016), the sizeof any population is correlated with

the level of fertility and mortality. It is positively correlated in the firstcase

and negatively correlated with the later. She argues that the Nigerian

population is growing at a high rate of2.5% per year due to a high fertility rate

and a decreasing mortality rate. Fapohundafurther argued that since the

3
population is growing rapidly due to a high fertility rate and a declining

mortality rate,an increasing proportion of thepopulation is under 15years of

age and is for the most economically inactive. Theyare basically dependent on

the economically active group.She thus maintained that as long as Nigerians

fertility remains high, the prospect is that the proportion of thepopulation who

are children will remain high and finally she maintained that the human

resources of a country aresimilar to the capital resources in that not only can

the stock increase through rapid birth rates, but investments canalso improve

its production capacities.

The Population and birth rate of a Nation cannot be overlooked when we talk

about Death rate. Death rates are important measurement of population

changes. They provide a way to compare health trend and other population

information from one year to another or from one community to anotherDeath

on the other hand, is caused basically by several diseases which are

environmental oriented. Poorenvironmental sanitations as evident in poor

sewage disposal system, inadequate supply of clean drinkable water aremajor

4
causes of some known diseases like Typhoid, Malaria, Cholera, and several

others.

Malaria is one of the commonest diseases that claims many live in tropical

countries like Nigeria. According toHanson (2017), malaria strikes an

estimated 100 million or more people in the developing countries each

year,killing one to two million of them. It is one of the biggest killers of

children in sub-Sahara Africa. The economic costof malaria is also stressed

by Anne Mills. According to Ann (2018), the impact of the disease on the

economy andeconomic development efforts as seen in reduced productivity

caused by workers illness and its financial burden onindividuals and

government cannot be over emphasized.Due to its widespread nature, precise

estimates of the mortality and morbidity of malaria are often hard to come

by.Unsurprisingly, there are no accurate statistical estimates of deaths caused

by malaria in Nigeria.

Jamison (2019), in his study in the 2019 world development report, attributed

several causes of death toenvironmental habit and social factors. According to

him, many cities suffer from air pollution caused by industries,power plants,

5
road transport and domestic use of coal. He stressed that atmospheric

pollution is a major cause ofrespiratorydisease in urban centres. Jamison

further stressed that inadequate environmentalsanitation and impurewater

supply are major causes of tuberculosis and typhoid diseases. He claims that

about1.3 billion people indeveloping countries lack access to clean and

portable water.

Briscoe and Whittington (2017), in their study on thedemand for water in

rural areas (with special reference to Nigeria and other five developing

countries) emphasizedthat improving the adequacy and quality of water

supply is a priority for rural development and governments andinternational

donors have an urgent role to play. Such efforts are already ongoing in Lagos

state by the United Nation(UN) as discussed in Prof. Johnson B. Falade’s

survey.

Other causes of deaths according to medical scientists arethose that arise from

habits such as tobacco, smoking,alcohol and illegal drugs use or drug abuse.

Althoughtobacco is in legal use everywhere in the world, but it causesfar

more deaths than all other “psycho-active” substances.An overview of deaths

6
such as this will never be complete if failure is made to mention those caused

by old age, war and accidents,especially accidents on roads, air and water.

1.2 HISTORICAL BACKGROUND OF THE CASE STUDY

ObafemiAwolowo University Teaching Hospitals Complex, was locate in Ile-

Ife town of Osun-State of west of Nigeria. It was established by the Federal

Government to provide qualitative health care delivery to its people. on the

year 1967, during the era of ObafemiAwolowo.

in 1967, the defunct western state government of Nigeria resolved to establish

a medical school in its state university at Ile-Ife (which was then five (5)

years old to provide manpower to tackle the health problem of the state. After

a period of careful planning Faculty of Health Science was created in the then

University of the (now OAU) on 8th May,1972.

On the 1st of July, 1975, the defunct western state Government established

the State Health Council a development in Nigeria. Health Management

System and created Ife/Ijesa Zone Health Board.

7
This hospital is a specialist hospital but they handle both maternity and

general services. It has chief genecology along with other medical and non-

medical staff, matron and nursing sisters.

This hospital also has various departments like

i. Medical statistics record section

ii. Laboratory section

iii. X-ray section

iv. Radiotherapy section

v. Physiotherapy section

vi. Station section and so on.

1.3 STATEMENT OF THE PROBLEM

The genesis of social problem is traced from birthand death rate which has an

effect on the socio-economic factors. This study seeks to find the rate of death

in ObafemiAwolowo Teaching Hospital Complex and the effect on social

economic factors.

8
1.4 RESEARCH QUESTIONS

i. What model can be used to evaluate the trend of death rate in OATHC?

ii. What is the seasonal effect on the rate of death occurrence in OATHC?

iii. What would be the future rate of death in OATHC?

1.5 AIM AND OBJECTIVES

The aim of this research is to estimate the trend of death rate in

ObafemiAwolowo Teaching Hospital Complex, Ile Ife and the objectives are

as follows;

i. to formulate Time series model for the rate of death in OATHC.

ii. to estimate the seasonal variation of the recorded cases of death in

OATHC.

iii. to predict the future rate of death in OATHC.

1.6 SIGNIFICANCE OF THE STUDY

This study will be significant to the hospital management, the state

government and the general public to know the rate at which people are

9
dying. This research document can also be used as a secondary source of

information to other researchers of similar importance that are willing to carry

out further research on this subject.

1.7 SCOPE AND LIMITATIONS

This research analyzed the data on annual death rate for the period of twenty

(20) years in ObafemiAwolowo Teaching Hospital Complex, Ile Ife.

The major limitation to this research work was basically on the collection of

data and information is regarded as confidential, besides there was an

inadequate access to the data due to reluctant attitude of the hospital medical

record department to release the data on time. As well as financial constraint.

1.8 DEFINITION OF TERMS

Below are the definitions of some terms used in this project for easy

comprehension.

i. Birth rate: This is the ratio between the number of live born births in

the year and the average total population of that year.

10
ii. Death rate: This is a measure of the number of deaths in a particular

population, scaled to the size of that population per unit of time.

iii. Death: this is the final cessation of vital functions or the end of life.

iv. Mortality: the process of diminution of a population because of death

or loss of life on large scale is known as mortality.

v. Mortality During Childhood: These measure death rates ending

infants less than four years.

vi. Mortality of Reproduction Ages: These measure death rates

associated with childbirth in the various ages groups in the

reproduction span of life between 15-49 years of age.

vii. Old Age Mortality Rate: This simply measures the rates of death at

the old age of various ages for sexes. The old age group is 70 years of

age and above.

viii.Rate: Frequency relative to a time unit.

ix. Analysis: This is the process of breaking a complex topic into smaller

parts in order to give a better understanding.

11
x. Time Series: It is a set of quantitative data which are obtained at a

regular interval of time.

xi. Forecasting: Prediction of future events and conditions are called

forecast, and the act of making such prediction is called forecasting.

xii. Time plot: There are basically four components of time series data

long term trend, circular, seasonal variation. To identify these

components, we need to plot the time series graph known as time plot

or histogram, it is a graph that plot the time series against its time

period.

xiii.Data: this is a set of values of qualitative or quantitative variables,

restated piece of information. Data as a general concept refers to the

fact that some existing information or knowledge is represented or

coded in some form suitable for better usage or processing.

12
 CHAPTER TWO

LITERATURE REVIEW

2.0 INTRODUCTION

In this chapter, we provide the most important knowledge to understand this

work. We first present a brief introduction of the time-series analysis methods

used in this project followed by study on related work.

The theoretical developments in time series analysis started early with

stochastic processes. The first actual application of autoregressive models to

data can be brought back to the work of G. U Yule and J. Walker in the 1920s

and 1930s. During this time the moving average was introduced to remove

periodic fluctuations in the time series, for example fluctuations due to

seasonality. Herman Wold introduced ARMA (AutoRegressive Moving

Average) models for stationary series, but was unable to derive a likelihood

function to enable maximum likelihood (ML) estimation of the parameters.

It took until 1970 before this was accomplished. At that time, the classic book

"Time Series Analysis" by G. E. P. Box and G. M. Jenkins came out,

13
containing the full modeling procedure for individual series: specification,

estimation, diagnostics and forecasting.

Nowadays, the so-called Box-Jenkins models are perhaps the most commonly

used and many techniques used for forecasting and seasonal adjustment can

be traced back to these models.

The first generalization was to accept multivariate ARMA models, among

which especially VAR models (Vector AutoRegressive) have become

popular. These techniques, however, are only applicable for stationary time

series. However, especially economic time series often exhibit a rising trend

suggesting non-stationarity, that is, a unit root.

Tests for unit roots developed mainly during the 1980s. In the multivariate

case, it was found that non-stationary time series could have a common unit

root. These time series are called cointegrated time series and can be used in

so called error-correction models within both long-term relationships and

short-term dynamics are estimated.

14
Another line of development in time series, originating from Box-Jenkins

models, are the non-linear generalizations, mainly ARCH (AutoRegressive

Conditional Heteroscedasticity) and GARCH (G = Generalized) models.

These models allow parameterization and prediction of non-constant variance.

These models have thus proved very useful for financial time series. The

invention of them and the launch of the error correction model gave C. W. J

Granger and R. F. Engle the Nobel Memorial Prize in Economic Sciences

in 2003.

Other non-linear models impose time-varying parameters or parameters

whose values changes when the process switches between different regimes.

These models have proved useful for modeling many macroeconomic time

series, which are widely considered to exhibit non-linear characteristics.

2.1 CONCEPTUAL REVIEW

Mortality rate or death rate is a measure of the number of deaths (in general,

or due to a specific cause) in a particular population, scaled to the size of that

15
population. Mortality rate is typically expressed in units of deaths per

1000 individuals per year.

Mortality rate is also measured per thousand. It is determined by how many

people of a certain age die per thousand people. Decrease of mortality rate is

one of the reasons for increase of population. Development of medical

science and other technologies has resulted in the decrease of mortality rate in

all the countries of the world for some decades. In 1990, the mortality rate of

children under 5 years of age was 144 per thousand, but in 2015 the child

mortality rate was 38 per thousand.

Perinatal mortality is the fetal deaths (stillbirths) past 22 or 28 completed

weeks of pregnancy. Maternal mortality is the number of deaths of mothers

assigned to pregnancy-related causes during a given time interval. Also,

Infant mortality is the number of deaths among children less than 1 year of

age during a given time interval. Child mortality is the number of deaths of

children less than 5 years old. Adult mortality is the number of deaths of adult

16
of age 18-60years. And old age mortality is the number of death of people

above 75years of age.

Ideally, all mortality estimation would be done using vital statistics and

census data. Census data will give detailed information about the population

at risk of death. The vital statistics provide information about live births and

deaths in the population. Often, either census data and vital statistics data is

not available. This is common in developing countries, countries that are in

conflict, areas where natural disasters have caused mass displacement, and

other areas where there is a humanitarian crisis

Scholars have stated that there is a significant relationship between a low

standard of living that results from low income and increased mortality rates.

A low standard of living is more likely to result in malnutrition, which can

make people more susceptible to disease and more likely to die from these

diseases. A lower standard of living may lead to as a lack of hygiene and

sanitation, increased exposure to and the spread of disease, and a lack of

access to proper medical care and facilities. Poor health can in turn contribute

17
to low and reduced incomes, which can create a loop known as the health-

poverty trap. Indian economist and philosopher AmartyaSen has stated that

mortality rates can serve as an indicator of economic success and failure.

2.2 EMPIRICAL AND THEORETICAL REVIEW

Behrman and Wolf (2015) used indirect methods to estimate levels and trends

of mortality in Malawi. The results indicated that source of drinking water

and sanitation facilities are strong predication of mortality.

Angaharan and Micheal (2017) used probit analysis to model mortality in

Pakistan and found that girls have a significant lower probability of dying in

age group 0 – 1 but have a significant higher probability of dying in the age

group 1 – 5. Thus, the higher mortality of girls in the age group 1 – 5 reflects

discrimination against girls in the form of lower health and other resources

inputs. Additionally, they found mother’s education beyond a certain

threshold and increased duration between births’ to significantly reduce child

mortality.

18
A study was done to describe the trend of child mortality in Kenya. It was

found that child mortality declined in Kenya in both rural and urban areas.

Authors concluded that there was a need of extra emphasis on urban slums to

reduce child mortality. (Murage et al, 2017)

Hussein (2017), used two models, one with the natural logarithmic

transformation of mortality rate time series and the other with successive

differences to provide mortality projections for the period 1997 – 2016 in

Egypt. Data were obtained from CAPMAS and UNICEF on the Egyptian

mortality rate for the period 1997 – 2016. The best model was determined by

successive steps of model specification, estimation, and comparison. Plots of

the data were provided for the original data for 1997 – 2016, the degree of

non-seasonal differencing, and a natural log transformation of the data. Plots

were also provided of the sample autocorrelation function and the sample

partial autocorrelation function for the original data, the degree of differences,

and the natural logarithmic transformations. The preferred model was an

autoregressive integrated moving average one for a first difference model

(model 1) and a natural logarithmic model (model 2). Parameter estimates in

19
model 2 were more significant and therefore preferred. Goodness of fit

comparisons and comparisons of plots of sample autocorrelation functions for

the errors with their probability limits showed both models to be adequate.

The two models were used to forecast mortality rate between 2017 and 2021.

Model 1 showed a faster decline in mortality than model 2: decline of 44.4%

compared to 25.9%. Model 2 results were preferred because of the known

inaccuracies in mortality data and the initially sharp decline between 1984

and 1985, which was due to implementation of government health programs.

Oronto (2015) argued that most of infant deaths are from preventable and

treatable condition, and almost all deaths are in poor countries. Infant and

child mortality rate vary among world regions and these differences are large

and increasing.

George et‘al (2016), stated that child deaths have a seasonal pattern occurring

more frequently during certain months of the year. There may exist

seasonality in death level among children, i.e there are more deaths occurring

in a particular time of the year or day due to specific diseases being rampant

20
in that particular period e.g. cases of death due to anemia, are predominant in

dry seasons when there is little vegetable, and also when malaria cases are

rampant causing breakdown of red blood cells. Cases due to malaria are most

predominant in the months of April, June, July, September, and December,

when the stagnant of water, which are used by mosquitoes as breading places.

Also, Jefferies (2016) in her study, she determined the relative significance of

environmental and maternal factors on childhood mortality in south western

Nigerian found that child mortality rate continued to be a function of an

environmental factor namely sources of drinking water and a child care

behavior factor, where the child was kept when the mother was at work.

Similarly, a Bayesian geo- additive survival models was introduced by

Adebayo (2018) to child mortality in Nigeria. The results showed the

existence of a district specific geographical variation in the level of child

mortality.

In a study from Kenya was done to examine relationship of rainfall and

temperature on malaria or anemia mortality. In this study, Malaria or anemia

21
mortality was found to be associated with changes in local temperature and

rainfall. It described the biological relationship as well. (Sewe et al, 2015)

In 1990Sub-Sahara Africa (SSA), there are 180 deaths per 1000 live births,

and only 9 deaths per 1000 live births in industrialized countries that is a 29-

fold differences in 2000, SSA had a mortality rate of 175 death per 1000

children and 6 deaths per 1000 children in industrialized countries, an

increase to 29- fold difference for the past three decades, significant progress

had been made towards the reduction of infant and child mortality rates in the

third world countries. As a result, between 1960 and 1993 in Arab states,

infant and child mortality deluded from 167 to 97 per 1000 live births. At

present, the high infant and child mortality rate in SSA is attracting

international researchers seeking effective and definitive health program or

methodology to implement sustainable measures or solution towards

reduction of infant and child mortality rate (Fagbeamibe and Alabi, 2018).

A study was conducted to describe the association between under-5 mortality

and socioeconomic, political, and healthcare factors between rich and poor

22
children. In this study, higher incomes were associated with lower under-5

mortality rates. (Houweling et al, 2015)

In a hospital based prospective study from Nigeria, Neonatal sepsis, birth

asphyxia, prematurity, IUGR were the common causes for admission.

Mortality rate, overall, was found to be 20.3%. Sepsis remained the major

cause of death followed by birth asphyxia and prematurity. (Ugwu, 2016)

23
 CHAPTER THREE

METHODOLOGY

3.0 INTRODUCTION

This chapter provides an overview of methodologies and methods that will be

utilized in this project. We start with discussing data collection and the

techniques applied in this work.

3.1 DATA COLLECTION

The data used in this project was a secondary data provided by

ObafemiAwolowo Teaching Hospital Complex Ile Ife. The dataset contains

accurate statistics time-series summary of Death cases report in the Hospital

over 20 years, which is very suitable for our project research.

3.2 METHOD OF DATA ANALYSIS

The goal of this study is to uncover the trend of death cases and develop an

effective model for projecting the future trend of death rates. The secondary

data for the research was produced by the hospital's records department for a

24
20-year period. This information was used to predict future death rates. Time

series Analysis was the statistical tool utilized to analyze the data in this

study.

3.3 TIME SERIES ANALYSIS

One of the primary goals of statistics is to estimate future levels of economic

activity by understanding historical data behavior. The study of time series is

one of the most essential tools for making predictions about the future based

on what has happened in the past. Which can be defined as a series of

observations made at predetermined intervals.

3.3.1 Uses of Time Series Analysis

The usage of time series models is;

(1) To obtain an understanding of the underlying forces and structure that

produced the observed data.

(2) To fit a model and proceed to forecasting, monitoring or even feedback

and feed forward control.

25
 (3) It is helpful for business planning as it helps in comparing the actual

current performance with the expected one

3.3.2 Advantages and Disadvantages of Time Series Analysis

i. Time Series Analysis helps to Identify Patterns. Memories are fragile

and prone to error.

ii. Time Series Analysis creates the Opportunity to Clean Data.

iii. Time Series Forecasting Can Predict the Future.

Time series analysis also suffers from a number of weaknesses, including

problems with generalization from a single study, difficulty in obtaining

appropriate measures, and problems with accurately identifying the correct

model to represent the data.

3.4 COMPONENT OF TIME SERIES ANALYSIS

The component of time series otherwise refers to as characteristics of time

series can be classified into 4 main types:

(1) Secular Trend (Tt)

26
 (2) Seasonal variation (St)

(3) Cyclical variation (Ct)

(4) Irregular movement (It)

TREND

The trend indicates the general tendency of the data to increase or decrease

over time. A trend is a continuous, long-term, average tendency. It is not

necessarily required for the rise or drop to be in the same direction during a

certain time period.

It is obvious that the tendency may rise, decrease, or remain consistent

throughout time. However, the overall tendency must be either upward,

downward, or stable. The population, agricultural productivity, products

manufactured, number of births and deaths, number of industry or any

factory, number of schools or colleges are some of its examples displaying

some kind of patterns of movement.

SEASONALVARIATIONS

This refers to a time series that appears to follow a similar or almost identical

pattern during corresponding months of succeeding years as a result of

27
primarily repeating events that occur on an annual basis. No matter how

frequently the data are collected—hourly, daily, weekly, quarterly, or monthly

—this variance will be visible in the time series.

Natural forces or customs created by humans both contribute to these

variances. Seasonal fluctuations are significantly influenced by the various

seasons or climatic conditions. Such as how crop production is influenced by

the seasons, how rainy season sales of umbrellas and raincoats spike, and how

summer sales of electric fans and air conditioning soar.

The impact of man-made conventions such as festivals, customs, habits,

trends, and special occasions such as marriage is plainly discernible. They

come back year after year. A seasonal rise should not be interpreted as a sign

of improved business circumstances.

CYCLICALVARIATIONS

Cyclical variations are those that manifest themselves in a time series over a

period longer than a year. The oscillation time for this movement is greater

than a year. A cycle is one whole period. This oscillation is commonly

referred to as the "business cycle."

28
The phases of prosperity, recession, depression, and recovery make up the

cycle's four stages. The cyclic variation may be periodic or irregular. Both

business upswings and downswings are influenced by the interplay and

unified character of the economic forces. Examples include variations in

temperature and rainfall on a daily basis, as well as some social and economic

factors..

IRREGULARVARIATIONS

This is used to describe the erratic or sporadic movement of time series

caused by a chance, random event that cannot be predicted and cannot be

directly separated. They lack determinism.

They are merely random or irregular variations, not regular variations. These

variations are chaotic, unanticipated, unmanageable, and unforeseeable. These

forces include natural calamities like earthquakes, wars, floods, and famines

as well as rapid changes in governmental regulations.

3.5 MATHEMATICAL MODEL FOR TIME SERIES ANALYSIS

In time series, there are two types of models that are usually considered as

good approximations to the time relationship among the components of the

29
observed data. The additive and multiplicative models are the most generally

assumed relationships between time series and their elements.

(1) Additive Model: This assumes that the value of composite series is

the sum of the four components, that is;

X t =T t + St +Ct + I t

(2) Multiplicative Model: This assumes that the value of composite

series is the product of four component values, that is;

X t =T t × St × Ct × I t

Where,

X t = Time series

T t = Value of secular Trend

St = Value of Seasonal variation

C t = Value of Cyclical Variation

I t = Value of Irregular variation

3.5.1 Measurement of Trend

The following are the four methods which are generally used for the study

and measurement of the trend component in a time series.

30
 i. The method of freehand,

ii. Least squares method.

iii. Moving average method

iv. Semi-average method.

In this research work, the method of moving average and the least square is

employed to estimate the trend value of time series by determine the equation

for the best time of fit and also use it to determine whether there will be an

increase or decrease in the observed data by predicting for the future

occurrence.

Least Square Method

This method is mostly employed and very satisfactory method to describe a

trend. It can be used to find equation of an appropriate trend line. This method

is similar to regression analysis.

Tt = a + bt

Where;

Tt = The dependent variable i.e the Trend

a = the intercept i.e the value of Yt when t=0

31
b = the slope i.e the change in Yt per unit change in time

t = the independent variable i.e time

a=∑
n
Xt
−b (∑ )
n
t

n ∑ t Xt −∑ t ∑ Xt
b=
n ∑ t −( ∑ t )
2 2

Moving Average Method

This is an average taking at the end of each successive period of time and is

related to the midpoint of an overall period.

The moving average method of estimating the trend is suitable when the

movement in the series is difficult to access. When the time plot does not

indicate the movement in the series, then we say the trend is obscured.

Let Xi denote theithmoving average of order n and Yi denote the series. It is

defined as ;

y 1 + y 2 + y 3 +…+ y n
X1=
n

y 2 + y 3 + y 4 + …+ y n + y n +1
X2=
n

32
 y 3 + y 4 + y 5+ …+ y n + y n +2
X3=
n

y N−n + y N−n +1+ y N −n +2 +…+ y N

X N−n =
n

Where N is the total number of series point in the data.

In this project work, the yearly data were used. Therefore, this is defined as:

y1 + y2 + y3 + y4
4 Years Moving Average ( 4 YMA )=
4

y1 M . A+ y2 M . A
4 Years MovingCentered Average ( 4 YMCA ) =
2

Method of Seasonal Index Average

i. Ratio to moving average: The trend is estimated by moving average

method and the ratio is taken as below;

X t =T t × St × Ct × I t

X t =T t × St ( ItandCtareignored )

Xt
St =
Tt

ii. Simple Average Method: The procedure for calculating seasonal

variation using simple average method is as follow:

a. Estimating the average value for each year of all years.

33
b. Compute year indices in the following respect

Average value of year × 4

Year Indices=
Total value of the year average

CHAPTER FOUR

34
 DATA PRESENTATION, ANALYSIS AND INTERPRETATION

4.0 INTRODUCTION

This section deals with the presentation, descriptive analysis and the time

series analysis of the quarterly record of death rate in OATHC, collected from

ObafemiAwolowo Teaching Hospital Complex, Ile Ife for the period of

20years.

4.1 PRESENTATION OF THE ORIGINAL DATA

Table I: Original Data Collected on Death Rate

The data below represents the data collected on quarterly record of death rate

from OATHC, for the period of 20 years (2003-2022).

1ST Quarter 2nd Quarter 3rd Quarter 4th Quarter Total

Year
M F M F M F M F M F

2003 28 18 75 56 62 43 68 51 233 168

2004 88 57 68 51 57 63 62 61 275 232

2005 80 49 39 39 63 44 59 46 241 178

2006 47 34 68 53 64 31 58 51 237 169

35
 2007 70 56 60 70 69 56 64 73 263 255

2008 54 40 81 58 62 66 77 59 274 223

2009 87 66 78 58 89 64 83 86 337 274

2010 80 53 69 64 56 47 84 68 289 232

2011 58 67 69 62 46 55 76 61 249 245

2012 57 62 84 88 92 58 102 50 335 258

2013 98 67 85 66 88 72 80 73 351 278

2014 86 77 71 87 57 34 24 22 238 220

2015 77 75 95 65 24 27 28 34 224 201

2016 39 50 92 62 103 77 91 46 325 235

2017 86 75 80 63 69 63 61 71 296 272

2018 74 54 73 60 51 41 70 68 268 223

2019 78 61 33 19 63 60 76 71 250 211

2020 86 62 69 55 72 75 67 75 294 267

2021 82 69 42 36 58 39 78 51 260 195

2022 51 58 63 63 40 28 61 54 215 203

Total 1406 1150 1394 1175 1285 1043 1369 1171 5454 4539

Source: OATHC Ile-Ife

Table II: Original Data (Male and Female Death Rate)

36
Year Quarter Male Female

2003 Q1 28 18

Q2 75 56

Q3 62 43

Q4 68 51

2004 Q1 88 57

Q2 68 51

Q3 57 63

Q4 62 61

2005 Q1 80 49

Q2 39 39

Q3 63 44

Q4 59 46

2006 Q1 47 34

Q2 68 53

Q3 64 31

Q4 58 51

2007 Q1 70 56

Q2 60 70

Q3 69 56

37
 Q4 64 73

2008 Q1 54 40

Q2 81 58

Q3 62 66

Q4 77 59

2009 Q1 87 66

Q2 78 58

Q3 89 64

Q4 83 86

2010 Q1 80 53

Q2 69 64

Q3 56 47

Q4 84 68

2011 Q1 58 67

Q2 69 62

Q3 46 55

Q4 76 61

2012 Q1 57 62

Q2 84 88

Q3 92 58

38
 Q4 102 50

2013 Q1 98 67

Q2 85 66

Q3 88 72

Q4 80 73

2014 Q1 86 77

Q2 71 87

Q3 57 34

Q4 24 22

2015 Q1 77 75

Q2 95 65

Q3 24 27

Q4 28 34

2016 Q1 39 50

Q2 92 62

Q3 103 77

Q4 91 46

2017 Q1 86 75

Q2 80 63

Q3 69 63

39
 Q4 61 71

2018 Q1 74 54

Q2 73 60

Q3 51 41

Q4 70 68

2019 Q1 78 61

Q2 33 19

Q3 63 60

Q4 76 71

2020 Q1 86 62

Q2 69 55

Q3 72 75

Q4 67 75

2021 Q1 82 69

Q2 42 36

Q3 58 39

Q4 78 51

2022 Q1 51 58

Q2 63 63

Q3 40 28

40
 Q4 61 54

Total 5454 4539

Source: OATHC Ile-Ife

Graph I: Time plot of Male and Female Death Rate

41
 Time plot of Male and Female Death rate
Male Female
120

100

80
Number of death

60

40

20

0
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
200 200 200 200 200 200 200 201 201 201 201 201 201 201 201 201 201 202 202 202
3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2
Year/Quarter

Interpretation: The graph above represents the time plot on quarterly record of male and

female death rate collected from OATHC Ile ife. The graph shows that there has been

fluctuation in death rate. It also reveals that male death is higher than female death over the

years. Therefore, the trend is not stable.

Table III: Original Data (Total Number of Death)

42
Year Quarter Total Death

2003 Q1 46

Q2 131

Q3 105

Q4 119

2004 Q1 145

Q2 119

Q3 120

Q4 123

2005 Q1 129

Q2 78

Q3 107

Q4 105

2006 Q1 81

Q2 121

Q3 95

Q4 109

2007 Q1 126

Q2 130

Q3 125

43
 Q4 137

2008 Q1 94

Q2 139

Q3 128

Q4 136

2009 Q1 153

Q2 136

Q3 153

Q4 169

2010 Q1 133

Q2 133

Q3 103

Q4 152

2011 Q1 125

Q2 131

Q3 101

Q4 137

2012 Q1 119

Q2 172

Q3 150

44
 Q4 152

2013 Q1 165

Q2 151

Q3 160

Q4 153

2014 Q1 163

Q2 158

Q3 91

Q4 46

2015 Q1 152

Q2 160

Q3 51

Q4 62

2016 Q1 89

Q2 154

Q3 180

Q4 137

2017 Q1 161

Q2 143

Q3 132

45
 Q4 132

2018 Q1 128

Q2 133

Q3 92

Q4 138

2019 Q1 139

Q2 52

Q3 123

Q4 147

2020 Q1 148

Q2 124

Q3 147

Q4 142

2021 Q1 151

Q2 78

Q3 97

Q4 129

2022 Q1 109

Q2 126

Q3 68

46
 Q4 115

Total 9993

Source: OATHC Ile-Ife

Graph II: Time plot of Total Number of Death

47
 Death Rate
200
180
160
140
120
Number of Death

100
80
60
40
20
0
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
200 200 200 200 200 200 200 201 201 201 201 201 201 201 201 201 201 202 202 202
3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2
Year/Quarter

Interpretation: Graph II above represents the time plot on quarterly record of

death rate collected from OATHC Ile-Ife which shows a fluctuating

movement. The graph indicates that there is low rate of death in the last

quarter of 2014 and the highest death rate that has ever been recorded in

OATHC is in the third quarter of 2016.

4.2 DESCRIPTIVE STATISTICS

48
Table IV: Descriptive Analysis of the Original Data on Male and Female Death Rate

The table below shows the descriptive analysis on death rate from 2003-2022.

Male Female

Parameter Value Parameter Value

Mean 68.175 Mean 56.7375
Standard Error 2.003618 Standard Error 1.701706
Median 69 Median 58.5
Mode 69 Mode 51
Standard Deviation 17.92091 Standard Deviation 15.22053
Sample Variance 321.1589 Sample Variance 231.6644
Kurtosis 0.069156 Kurtosis 0.142519
Skewness -0.52839 Skewness -0.50723
Range 79 Range 70
Minimum 24 Minimum 18
Maximum 103 Maximum 88
Sum 5454 Sum 4539
Count 80 Count 80
Source: SPSS 22

Interpretation:Table 3 above shows the descriptive analysis output on male

and female death rate for 20years, it can be deduced that over this period, the

average number of death recorded per quarter on both male and female are

49
68.175(68 death approximately) and 56.74(57 death approximately)

respectively an associate standard error of 2.004 and 1.702.

The least number of deaths that has ever been recorded per quarter on both

male and female are 24 and 18 deaths while the highest number of deaths

recorded are 103 and 88 respectively.

Table V: Descriptive Analysis of the Original Data on Death Rate

Death Rate

Parameter Value
Mean 124.9125
Standard Error 3.389027
Median 130.5
Mode 119
Standard Deviation 30.31238
Sample Variance 918.8403
Kurtosis 0.412291
Skewness -0.8544
Range 134
Minimum 46
Maximum 180
Sum 9993
Count 80

50
Source: SPSS 22

Interpretation: The table above shows the descriptive analysis output on

death rate for 20years, it can be deduced from the table that

124.913(approximately 125 people) die averagely per quarter with an

associate standard error of 3.389.

The minimum death rate that has ever been recorded over the years is 46 and

the maximum death rate that has ever been recorded over the years is 180.

According to the kurtosis analysis, it indicates that the distribution is

positively and normally peaked (0.4123). Also, the skewness analysis shows

that the distribution is moderately skewed (-0.8544).

4.3 TIME SERIES ANALYSIS

51
Table VI: Estimation of Trend Equation using Least Square Method

Year Q Xt Q
2
Q Xt T =a+ bq

2003 1 46 1 46 121.9071

2 131 4 262 121.9842

3 105 9 315 122.0613

4 119 16 476 122.1384

2004 5 145 25 725 122.2156

6 119 36 714 122.2927

7 120 49 840 122.3698

8 123 64 984 122.4469

2005 9 129 81 1161 122.524

10 78 100 780 122.6011

11 107 121 1177 122.6782

12 105 144 1260 122.7553

2006 13 81 169 1053 122.8324

14 121 196 1694 122.9095

52
 15 95 225 1425 122.9867

16 109 256 1744 123.0638

2007 17 126 289 2142 123.1409

18 130 324 2340 123.218

19 125 361 2375 123.2951

20 137 400 2740 123.3722

2008 21 94 441 1974 123.4493

22 139 484 3058 123.5264

23 128 529 2944 123.6035

24 136 576 3264 123.6806

2009 25 153 625 3825 123.7578

26 136 676 3536 123.8349

27 153 729 4131 123.912

28 169 784 4732 123.9891

2010 29 133 841 3857 124.0662

53
 30 133 900 3990 124.1433

31 103 961 3193 124.2204

32 152 1024 4864 124.2975

2011 33 125 1089 4125 124.3746

34 131 1156 4454 124.4517

35 101 1225 3535 124.5289

36 137 1296 4932 124.606

2012 37 119 1369 4403 124.6831

38 172 1444 6536 124.7602

39 150 1521 5850 124.8373

40 152 1600 6080 124.9144

2013 41 165 1681 6765 124.9915

42 151 1764 6342 125.0686

43 160 1849 6880 125.1457

44 153 1936 6732 125.2228

54
2014 45 163 2025 7335 125.3

46 158 2116 7268 125.3771

47 91 2209 4277 125.4542

48 46 2304 2208 125.5313

2015 49 152 2401 7448 125.6084

50 160 2500 8000 125.6855

51 51 2601 2601 125.7626

52 62 2704 3224 125.8397

2016 53 89 2809 4717 125.9168

54 154 2916 8316 125.9939

55 180 3025 9900 126.0711

56 137 3136 7672 126.1482

2017 57 161 3249 9177 126.2253

58 143 3364 8294 126.3024

59 132 3481 7788 126.3795

55
 60 132 3600 7920 126.4566

2018 61 128 3721 7808 126.5337

62 133 3844 8246 126.6108

63 92 3969 5796 126.6879

64 138 4096 8832 126.765

2019 65 139 4225 9035 126.8422

66 52 4356 3432 126.9193

67 123 4489 8241 126.9964

68 147 4624 9996 127.0735

2020 69 148 4761 10212 127.1506

70 124 4900 8680 127.2277

71 147 5041 10437 127.3048

72 142 5184 10224 127.3819

2021 73 151 5329 11023 127.459

74 78 5476 5772 127.5361

56
 75 97 5625 7275 127.6133

76 129 5776 9804 127.6904

2022 77 109 5929 8393 127.7675

78 126 6084 9828 127.8446

79 68 6241 5372 127.9217

80 115 6400 9200 127.9988

Total 3240 9993 173880 408006

Source: SPSS 22

∑ Q=3240∑ Xt=9993

∑ QXt =408006

∑ Q 2=173880
n=80

The trend equation is Tt =a+ b(Q)

57
a=∑
n
Xt
−b (∑ )
n
Q

n ∑ QXt−∑ Q ∑ Xt
b=
n ∑ Q 2− ( ∑ Q )
2

80 ( 408006 )−(3240 × 9993)

b=
80 ( 173880 )−¿ ¿

32640480−32377320
b = 13910400−10497600

263160
b = 3412800

b = 0.07711

9996 3240
a = 80 −(0.07711 × 80 )

a = 124.95−0.07711(40.5)

a = 124.95−3.122955

a = 121.83

The Trend equation is now; Tt =121.83+0.07711(Q)

58
Graph III: Graphical representation of Estimation of trend using Least Square

method.

Estimation of trend using Least Square method

130

128

126

124

122

120

118
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
200 200 200 200 200 200 200 201 201 201 201 201 201 201 201 201 201 202 202 202
3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2

Interpretation: The above graph represents the estimation of trend using least square

method, which shows there is an upward trend movement i.e the rate of death is slightly

increasing in OAUTHC.

59
Table VII: Estimation of Trend using 4 Quarters Moving Average

Year Q Xt 4 Q MA 4 QCMA (Trend)

2003 1 46

2 131

3 105 100.25 112.625

4 119 125 123.5

2004 5 145 122 123.875

6 119 125.75 126.25

7 120 126.75 124.75

8 123 122.75 117.625

2005 9 129 112.5 110.875

10 78 109.25 107

11 107 104.75 98.75

12 105 92.75 98.125

2006 13 81 103.5 102

60
 14 121 100.5 101

15 95 101.5 107.125

16 109 112.75 113.875

2007 17 126 115 118.75

18 130 122.5 126

19 125 129.5 125.5

20 137 121.5 122.625

2008 21 94 123.75 124.125

22 139 124.5 124.375

23 128 124.25 131.625

24 136 139 138.625

2009 25 153 138.25 141.375

26 136 144.5 148.625

27 153 152.75 150.25

28 169 147.75 147.375

2010 29 133 147 140.75

61
 30 133 134.5 132.375

31 103 130.25 129.25

32 152 128.25 128

2011 33 125 127.75 127.5

34 131 127.25 125.375

35 101 123.5 122.75

36 137 122 127.125

2012 37 119 132.25 138.375

38 172 144.5 146.375

39 150 148.25 154

40 152 159.75 157.125

2013 41 165 154.5 155.75

42 151 157 157.125

43 160 157.25 157

62
 44 153 156.75 157.625

2014 45 163 158.5 149.875

46 158 141.25 127.875

47 91 114.5 113.125

48 46 111.75 112

2015 49 152 112.25 107.25

50 160 102.25 104.25

51 51 106.25 98.375

52 62 90.5 89.75

2016 53 89 89 105.125

54 154 121.25 130.625

55 180 140 149

56 137 158 156.625

2017 57 161 155.25 149.25

58 143 143.25 142.625

63
 59 132 142 137.875

60 132 133.75 132.5

2018 61 128 131.25 126.25

62 133 121.25 122

63 92 122.75 124.125

64 138 125.5 115.375

2019 65 139 105.25 109.125

66 52 113 114.125

67 123 115.25 116.375

68 147 117.5 126.5

2020 69 148 135.5 138.5

70 124 141.5 140.875

71 147 140.25 140.625

72 142 141 135.25

2021 73 151 129.5 123.25

64
 74 78 117 115.375

75 97 113.75 108.5

76 129 103.25 109.25

2022 77 109 115.25 111.625

78 126 108 106.25

79 68 104.5

80 115

Source: SPSS 22

65
Graph IV: Graphical representation of Estimation of trend using 4quarters moving

average.

Estimation of trend using Moving Average

180

160

140

120

100
Trend

80

60

40

20

0
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
200 200 200 200 200 200 200 201 201 201 201 201 201 201 201 201 201 202 202 202
3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2
Year/Quarter

Interpretation: The graph above represents the estimation of trend using

moving average. It can be deduced from the graph that the rate of death is

fluctuating with slight upward trend movement.

66
Table VIII: Estimation of Seasonal Variation using 4QCMA Trend Value.

Xt
St =
YEAR Q Xt Tt Tt

2003 1 46

2 131

3 105 112.625 0.932297

4 119 123.5 0.963563

2004 5 145 123.875 1.170535

6 119 126.25 0.942574

7 120 124.75 0.961924

8 123 117.625 1.045696

2005 9 129 110.875 1.163472

10 78 107 0.728972

11 107 98.75 1.083544

12 105 98.125 1.070064

2006 13 81 102 0.794118

67
 14 121 101 1.19802

15 95 107.125 0.886814

16 109 113.875 0.95719

2007 17 126 118.75 1.061053

18 130 126 1.031746

19 125 125.5 0.996016

20 137 122.625 1.117227

2008 21 94 124.125 0.757301

22 139 124.375 1.117588

23 128 131.625 0.97246

24 136 138.625 0.981064

2009 25 153 141.375 1.082228

26 136 148.625 0.915055

27 153 150.25 1.018303

28 169 147.375 1.146735

2010 29 133 140.75 0.944938

68
 30 133 132.375 1.004721

31 103 129.25 0.796905

32 152 128 1.1875

2011 33 125 127.5 0.980392

34 131 125.375 1.044865

35 101 122.75 0.822811

36 137 127.125 1.077679

2012 37 119 138.375 0.859982

38 172 146.375 1.175064

39 150 154 0.974026

40 152 157.125 0.967383

2013 41 165 155.75 1.05939

42 151 157.125 0.961018

43 160 157 1.019108

69
 44 153 157.625 0.970658

2014 45 163 149.875 1.087573

46 158 127.875 1.235582

47 91 113.125 0.80442

48 46 112 0.410714

2015 49 152 107.25 1.417249

50 160 104.25 1.534772

51 51 98.375 0.518424

52 62 89.75 0.690808

2016 53 89 105.125 0.846611

54 154 130.625 1.178947

55 180 149 1.208054

56 137 156.625 0.874701

2017 57 161 149.25 1.078727

58 143 142.625 1.002629

70
 59 132 137.875 0.957389

60 132 132.5 0.996226

2018 61 128 126.25 1.013861

62 133 122 1.090164

63 92 124.125 0.741188

64 138 115.375 1.1961

2019 65 139 109.125 1.273769

66 52 114.125 0.455641

67 123 116.375 1.056928

68 147 126.5 1.162055

2020 69 148 138.5 1.068592

70 124 140.875 0.880213

71 147 140.625 1.045333

72 142 135.25 1.049908

2021 73 151 123.25 1.225152

71
 74 78 115.375 0.676056

75 97 108.5 0.894009

76 129 109.25 1.180778

2022 77 109 111.625 0.976484

78 126 106.25 1.185882

79 68

80 115

Graph V: Graphical representation of Estimation of Seasonal Variation.

72
 Estimation of Seasonal Variation
1.8

1.6

1.4

1.2

1
Variation

0.8

0.6

0.4

0.2

0
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
200 200 200 200 200 200 200 201 201 201 201 201 201 201 201 201 201 202 202 202
3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2
Year/Quarter

Interpretation: The above graph shows the graphical representation of

seasonal variation, which shows a fluctuating trend movement

Table IX: Computation of Seasonal Index

73
Year Q1 Q2 Q3 Q4 TOTAL

2003 0.932297 0.963563

2004 1.170535 0.942574 0.961924 1.045696

2005 1.163472 0.728972 1.083544 1.070064

2006 0.794118 1.19802 0.886814 0.95719

2007 1.061053 1.031746 0.996016 1.117227

2008 0.757301 1.117588 0.97246 0.981064

2009 1.082228 0.915055 1.018303 1.146735

2010 0.944938 1.004721 0.796905 1.1875

2011 0.980392 1.044865 0.822811 1.077679

2012 1.077679 0.859982 0.974026 0.967383

2013 1.05939 0.961018 1.019108 0.970658

2014 1.087573 1.235582 0.80442 0.410714

2015 1.417249 1.534772 0.518424 0.690808

2016 0.846611 1.178947 1.208054 0.874701

74
 2017 1.078727 1.002629 0.957389 0.996226

2018 1.013861 1.090164 0.741188 1.1961

2019 1.273769 0.455641 1.056928 1.162055

2020 1.068592 0.880213 1.045333 1.049908

2021 1.225152 0.676056 0.894009 1.180778

2022 0.976484 1.185882

QT 20.079124 19.04443 17.689953 19.046049

QM 1.056796 1.002338 0.93105016 1.00242363 3.99260805

ADJUSTED

(QM x 1.05875256 1.004194 0.93277391 1.00427953

4)/QMT

S.I 1.059 1.004 0.933 1.004 4

Table X: Forecast for year 2022-2025 using Least Square Method

75
 FORECAST =¿

Year Q Tt =a+ b(Q) S.I (Trend × SI )

128.0759 1.059 135.6324

2023 81

128.153 1.004 128.6656

82

128.2301 0.933 119.6387

83

128.3072 1.004 128.8205

84

128.3844 1.059 135.959

2024 85

128.4615 1.004 128.9753

86

128.5386 0.933 119.9265

87

128.6157 1.004 129.1301

88

128.6928 1.059 136.2857

2025 89

128.7699 1.004 129.285

90

128.847 0.933 120.2143

91

128.9241 1.004 129.4398

92

Graph VI: Graphical representation of Forecast.

76
 Forecast
140

135

130
Trend Value

125

120

115

110
Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4
2023 2024 2025
Year/Quarter

Interpretation: The graph above shows the graphical representation of

Forecast from 2023 to 2025 quarterly. It can be deduced from the graph that

the fluctuating trend movement in death rate would continue over the years.

CHAPTER FIVE

77
 FINDINGS, CONCLUSION AND RECOMMENDATION

5.0 INTRODUCTION

This chapter presents the findings, conclusion and recommendation drawn

from the analysis observed in the previous chapter.

5.1 FINDINGS

Twenty (20) years data were extracted from the record document of

ObafemiAwolowo Teaching Hospital Complex Il-Ife on number of death, the

data were obtained on quarterly basis. Time series analysis was used to

determine the trend` line and also used to fit the linear trend model; the model

was used to predict for future occurrence on death rate.

Based the analysis conducted in the chapter four, the analysis is in two parts.

The first part contains the descriptive analysis which has the estimate of mean

and its standard error, skewness etc. And the second part contains the Time

series Analysis which has the estimate of trend using least square method,

moving average also the seasonal variation with the appropriate time plots.

78
From the output of the descriptive analysis on male and female death rate, it

can be deduced that the average number of death recorded per quarter on both

male and female are 68 and 57 deaths respectively with an associate standard

error of 2.004 and 1.702. And the least number of deaths that has ever been

recorded per quarter on both male and female are 24 and 18 deaths while the

highest number of deaths recorded are 103 and 88 respectively.

Also, from the output of the descriptive analysis on total death rate, it can be

deduced that over the period of 20 years, 125 people die averagely per quarter

with an associate standard error of 3.389.

The minimum death rate that has ever been recorded over the years is 46 and

the maximum death rate that has ever been recorded over the years is 180.

According to the kurtosis analysis, it indicates that the distribution is

positively and normally peaked (0.4123). Also, the skewness analysis shows

that the distribution is moderately skewed (-0.8544).

79
The time plots for the original data shows that the rate of death is fluctuating

over the years with a slight upward trend movement. It was observed that the

Trend equation Tt =121.83+0.07711(Q)which implies that the average

occurrence death is 122 and the rate at which the occurrence is increasing is

by 0.07711.

The Trend equation was used to forecast for year 2023-2025 and it was

discovered that there will be fluctuation in the number of death occurring in

the future.

5.2 CONCLUSION

The results of this study indicated a fluctuating and a slight upward trend in

the number of death over the period under study.

The results of this study show that incidence of death is affected or influenced

by seasonal factors. The identified models for death rate were used to forecast

future occurrence taking into consideration the seasonal factor and trend

influencing the occurrence of death in the hospital. The forecast made in this

study shows that the fluctuation in the number of death will continue.

80
5.3. RECOMMENDATION

In other to eradicate the slight high rate of death, below are the possible ways

to reduce the death rate in the state and perhaps in the nation as a whole.

Some factors have to be strictly put into consideration in other to prevent

future increase in death rate.

 Each government-owned hospital should have access to adequate

medical facilities and employees. The facilities must be set up so that

they can supply or render the greatest amount of medical care for the

population that is connected to the institution (hospital).

 The state's health sector needs to be adequately funded if it is to

provide for the population of the state or the residents of a particular

locality to have sound and good health.

 It was noted that many of our roads which some commuters now refer

to as "death traps" saw the loss of numerous lives. A significant number

of fatalities occur each year as a result of inadequate supplies of

routinely drinking water. Government should therefore offer enough

81
 social facilities so that at least a typical person can enjoy life as a

citizen.

 The necessary social infrastructure should be offered.

 If Nigeria is to reach her highest degree of health standard, the federal

government, privately held medical facilities, and every citizen must

take care of their various parts so that the death rate would be at a low

level in our society

REFERENCES

82
Adebayo Y. (2018): “Preventive healthcare of infants in a region of Lebanon:

parental beliefs, attitudes and behaviors.” Maternal Child Health

Journal. 9 (1): 83-90.

Behrman J. and Wolfe B. (2015): “More evidence on nutritional demand:

Income seems overrated and women’s schooling underemphasized.”

Journal of Development Economy 14.105-128.

Fagbeamibe, A &Alabi, O (2018): “Differentials and Correlates of Infant

Mortality in Nigeria; A Comparative Survival Analysis Between North-

East and South-West Nigeria.” International Journal of Tropical

Disease and Health,4 Retrieved; August, 2018.

Gangaharm C. & Michael R. (2017): “Inequality and Infant and Childhood

Mortality in the United States in the Twentieth Century”. NBER

Working Paper No. 16133.

George et al (2016): “Times Series Analysis Forecasting and Control.” (3rd

edition), Prentice Hall Inc. New Jersey, USA. 16-19

83
Houweling et al (2015): “Under-5 mortality pattern among Nigerian Children

in the 2015”. Journal of Tropical Sciences.pg1(3):4-6

Hussein 2017: “The natural logarithmic transformation of mortality rate time

series and the other with successive differences to provide mortality

projections for the period 1997 – 2016 in Egypt: A Statistical Analysis.

Journal of Applied Quantitative Method, Vol. 4, No. 3, 2017

Jefferies J. (2016): “Fertility assumptions for the 2016-based National

populations. Population trends.131: 19-27

Murage et al(2017): “The trend of child mortality in Kenya” A Population-

based Analysis of Individual and Community-Level Determinants.

Population Health Metrics, Vol. 9, No. 6, 2011.

Oronto D. (2015): “Global Strategy for Reversal of High Infant Mortality

Rate in Nigeria Sub-Sahara Africa.” Journal of Environmental and

Disaster. 8(3) 43-56

Sewe et al (2015): “Rainfall and Temperature on malaria or anemia mortality

in Kenya” Maternal Child Health Journal. 9 (1): 83-90

84
Ugwu 2016: “The major cause of death, birth asphyxia and prematurity in

Nigeria” International Journal of Scientific & Technology Research,

Vol. 2, Issue 11, November 2016. P73-78.ISSN 2277-8616.

WHO (2021): “Global Health Observatory Data; Health Outcomes in Urban

population”. Retrieved from http://www.who.gho.com 17th January,

2022.

85