A MULTIPLE REGRESSION ANALYSIS OF GOVERNMENT RECURRENT EXPENDITURE AND ITS COMPONENT IN NIGERIA

A MULTIPLE REGRESSION ANALYSIS OF GOVERNMENT    RECURRENT EXPENDITURE AND ITS COMPONENT IN NIGERIA

BY



ABDULLAHI ABUBAKAR



ADM.NO. 1310212025



PROJECT SUBMITTED TO THE DEPARTMENT OF STATISTICS FACULTY OF SCIENCE, KEBBI STATE UNIVERSITY OF SCIENCE AND TECHNOLOGY, ALIERO.
IN PARTIAL FULFILMENT OF THE REQUIREMENT FOR THE AWARD OF A BACHELOR OF SCIENCE (B.SC HONS) DEGREE IN STATISTICS


Approval
This is to declare that this research carried out by Abdullahi Abubakar with admission number (1310212025) of mathematics department (statistics unit), is fully adequate in scope and qualify for the award of Degree of Bachelor of Science (B.sc Hon) in statistics. From Kebbi State University of Science and Technology, Aliero
   


-----------------------------------                                                        -----------------------------------                                                                                                                                                     
Dr.  N.O. Nweze                                                                                         date
(Project supervisor)








-----------------------------------                                                        -----------------------------------
Dr R.V. Singh                                                                                              date
(Head of department)









-----------------------------------                                                        -----------------------------------
External Examiner                                                                                        date



Dedication
I dedicate this project to God Almighty my creator, my strong pillar, my source of inspiration, wisdom, knowledge and understanding. He has been the source of my strength throughout this program and on His wings only have I soared. I also dedicate this work to my humble mother; mal.Ma’inna Ajah, and my beloved father mal. Abdullahi Aliyu who has encouraged me all the way and whose encouragement has made sure that I give it all it takes to finish that which I have started. May Almighty Allah in his infinite mercy bless them abundantly (ameen).
















Acknowledgment
I would like to express my special thanks of gratitude to my teacher (DR. N.O. Nweze) as well as the head of department (DR. R.V. Singh) who gave me the golden opportunity to do this wonderful project on the topic (A Multiple regression Analysis of Government Recurrent Expenditure and it’s Component in Nigeria), which also helped me in doing a lot of Research and I came to know about so many new things I am really thankful to them.
Secondly I would also like to thank my parents and friends who helped me a lot in finalizing this project within the limited time frame.















ABSTRACT
This research presents the multiple regression analysis of government recurrent expenditure and its component in Nigeria. Three expenditures were used namely: Health, Agriculture and Education were considered as predictor variables while recurrent expenditure was taken as response. Time series data for the period 2005 to 2017, was used in the study .Multiple Linear Regression was used as statistical tool , Also the variables were tested using analysis of variance and correlation was also performed. The study finds that Health, Agriculture and Education are positively influenced by Total Recurrent Expenditure. While analysis of variance reveals that there is linear relationship between the Total Recurrent Expenditure and its components. Moreover scatter plots were drawn to show the kind of relationships existing between recurrent expenditure and each of the three components. Before conducting ANOVA, the data sets were inspected for normality assumption using normal probability plot of the standardized residual in the mini Tab. The plot revealed that normality assumption was not violated. And multicollinearity problems were further inspected using Durbin-Watson and VIF, and it was discovered that the problems existed in the data sets. After the application of All Possible Regression Method of Selection of the Best Regression Model, Health Expenditure was recommended as the principal determinant of the recurrent expenditure in Nigeria. The researches recommended that allocation government spending need to be bas on the level needs and versatility of individual sectors.











TABLE OF CONTENTS
Title page ………………......…………………………………………………………………….i
Approval page …………...………………………………………………………………...…....ii
Dedication ………………...……………………………………………………………….…....iii
Acknowledgment,,,,,,,,,,,,,,,,, ……………………………………………………………….….iv
Abstract ……………………..…………………………………………………………………...v
CHAPTER ONE
1.1 Introduction…………... …………………………………………………………….......1
1.2 Aim and Objectives…………... ………………………………………………………..2
1.3 Significance of the study …………….…………………………………………………2
1.4 Scope of the Study…………………...…………………………………………………3
1.5 Research hypothesis ………………...………………………………………………...3
1.6 Definition of relevance terms…………… …………………………………………….4
CHAPTER TWO
2.1 literature review …………………………..……………………………………….........5
2.2 Significance of Health, Agriculture and Education in Nigeria………………..…..…7

CHAPTER THREE
3.1 Source of data …………………………………………………………………..……...9
3.2 Simple linear regression ………………………………………………………...……..9
3.3 Multiple linear regression ………………………………………………………..…...10
3.4 Least Square Method………………………………………………………………....10
3.5 Derivation of normal equations………………….…………………………………...11
3.6 Measure of Linear Correlation ……………….………………………………………14
3.7 Hypothesis test in Multiple Linear Regression ……………….…………………....14
3.8 Estimating    ………………………………………………………………………...15
3.9 Coefficient of multiple determination…………………………….…………………..15
3.1.0 Test for multicollinearity ………………………………………………………….......15
3.1.1 Discussion of the result ………………………………………….………………...…16
CHAPTER FOUR
4.1 Summary ………………………………………………………….……………………25
4.2 Conclusion ………………………………………………………..……………………25
4.3 Recommendation …………………………………………….....………………….…25
References …………………………………………………….………………………………27














CHAPTER ONE
INTRODUCTION
1.1 Background of the study
Government expenditures play key roles in the operation of all economies. It refers to expenses incurred by the government for the maintenance of itself and provision of public goods, services and works needed to foster or promote economic growth and improve the welfare of people in the society. Government (public) expenditures are generally categorized into expenditures on administration, defense internal securities, health, education, foreign affairs, etc. and have both capital and recurrent components. Capital expenditure refers to the amount spent in the acquisition of fixed (productive) assets (whose useful life extends beyond the accounting or fiscal year), as well as expenditure incurred in the upgrade/improvement of existing fixed assets such as lands, building, roads, machines and equipment, etc., including intangible assets. Expenditure in research also falls within this component of government expenditure. Capital expenditure is usually seen as expenditure creating future benefits, as there could be some lags between when it is incurred and when it takes effect on the economy. Recurrent expenditure on the other hand refers to expenditure on purchase of goods and services, wages and salaries, operations as well as current grants and subsidies (usually classified as transfer payments). Recurrent expenditure, excluding transfer payments, is also referred to as government final consumption expenditure. The annual budget spells out the direction of the expected expenditure, as it contains details of the proposed expenditure for each year, though the actual expenditures may differ from the budget figures due, for example, to extra-budgetary expenditures or allocations during the course of the fiscal year.
Government expenditure is a major component of national income as seen in the expenditure approach to measuring national income: (Y = C+I+G +(X – M)). This implies that government expenditure is a key determinant of the size of the economy and of economic growth. However, it could act as a two-edged sword: It could significantly boost aggregate output, especially in developing countries where there are massive market failures and poverty traps, and it could also have adverse consequences such as unintended inflation and boom-bust cycles (Wang and Wen, 2013). The effectiveness of government expenditure in expanding the economy and fostering rapid economic growth depends on whether it is productive or unproductive. All things being equal, productive government expenditure would have positive effect on the economy, while unproductive expenditure would have the reverse effect
1.2 Aim and objectives of the study
1.2.1 Aim
The aim of this study is to know if there is correlation between the Health Agriculture and Education in Recurrent Expenditure (T R E).
1.2.2 Objectives

To educate Nigerians if there is significant relationship between government expenditure (Health, Agriculture and Education) on total recurrent expenditure.
By measuring the difference between the expenditure.
To determine the suitable model connecting Health, Agriculture and Education with total recurrent expenditure.
To determine different in budgetary allocations to the Nigerian economic sectors  and  Recurrent Expenditure from 2005 to 2017;

1.3 significance of the study
The study attempted to examine empirically the structure and growth of the federal government expenditure in Nigeria. Over the years, growth in public expenditure has continued to generate debates in as to what are the determinants of government expenditure growth in Nigeria. Thus, the main purpose of this study was to investigate what factors cause the growth in public expenditure growth in Nigeria.

1.4 Scope of the study
The scope of this study is centered on the analysis of government expenditure in Nigeria from 2000-1014 in essence the study employs multiple regression analysis in the apprehension of the problem. Even through, a considerable size of population need to be observed in other to come up with a better picture of the problem, the study has chosen a sample of 13 years analyze.
Mostly data used are secondary data extracted from statistical publication and are not as reliable as primary data.

1.5 Research hypotheses
Hypothesis 1
   :   There is no relationship between Recurrent Expenditure and Health
   :   There is relationship between Recurrent Expenditure and Health

Hypothesis 2
   :   There is no relationship between Recurrent Expenditure and Agriculture
   :   There is relationship between Recurrent Expenditure and Agriculture

Hypothesis 3
   :   There is no relationship between Recurrent Expenditure and Education
   :   There is relationship between Recurrent Expenditure and Education

Level of significance: 

Decision rule   : Reject  if  is less than the Level of significance ( and accept  if otherwise.




1.7 Definition of relevant terms
Recurrent expenditure
Recurrent expenditure refers to payments made by governments or organizations for all purposes except capital costs. Recurrent expenditure includes payments made on goods and services as well as interest and subsidies.
Health
Health is the level of functional and metabolic efficiency of a living organism.
Agriculture
Is the production of plants and animals useful to man.
 Education
Education is the process of facilitating learning, or the acquisition of knowledge, skills, values, beliefs, and habits.

CHAPTER TWO
LITERATURE REVIEW

The determinants of government expenditures are well documented in the extant literature. Economic theories and hypotheses such as those of Wagner, Wiseman and Peacock, etc. help to explain the determinants of the growth of government expenditure, prominent among which is the size of the government. Specifically, Wagner’s law (also referred as the law of increasing State activity or the law of expanding State role) states that as the economy develops (evidenced in high rate of industrialization and the growth of per capita income), the share of government expenditure in the gross national product tends to rise accordingly. Here, the growth of government expenditure is attributed to economic growth and development. Peacock and Wagner’s hypothesis emanated from a study that was premised on Wagner’s Law. According to this hypothesis, industrialization which elicits increased government spending also enhances government revenue generation, particularly through taxation which is used to finance government expenditure. Peacock and Wiseman were however of the view that government expenditure evolves in a step-like pattern, owing to variations in pattern of government expenditure in periods of upheaval and periods of relative calmness. Government revenue from taxation increases in period of upheaval as the tax-resistance level of the people tends to decline. The revenue generated from enhanced taxation is used to finance government expenditure which expectedly increases during the period of upheaval. Once calmness is restored, government expenditure does not usually go down to its previous trend. 
The Keynesian theory asserts that government expenditure especially deficit financing could provide short - term stimulus to help halt a recession or depression. The Keynesians however advised that policy makers should be prepared to reduce government expenditure once the economy recovers to forestall inflation (Mitchell, 2005).  Much empirical researches have been conducted to investigate the impact of government expenditure on economic growth in various countries. The results however have been mixed. While some observe that public expenditure favours growth, others argue that excessive government expenditure could be detrimental to growth.  Sinha (1998) studies the relationship between government expenditure and GDP in China, and finds that a strong positive relationship exists between both variables. The Granger causality test shows there is evidence (though weak) of unidirectional causality, with causality running from government expenditure to GDP. Loizides and Vamvoukas (2004) investigate the causal relationship between the relative size of government (measured as the share of total expenditure in GNP) and economic growth rate using data on Greece, UK and Ireland, and find that government size Granger causes economic growth rate in all three countries in the short run, and in the long run for Ireland and UK, and that economic growth Granger causes increase in the relative size of government in Greece, and Ireland when inflation is included.
In a study to examine the growth effects of public expenditure for a panel of 30 developing countries over the 1970s and 1980s Bose et al (2007) finds that the share of government capital expenditure in GDP is positively and significantly correlated with economic growth, while current expenditure is observed to be insignificant. At the disaggregated level, government investment in education and total expenditures in education are the only outlays that were observed to be significantly associated with growth it the budget constraint and omitted variables are taken into consideration. Applying two different panel data methodologies to seven transition economies in South Eastern Europe, Alexiou (2009) finds evidence for the support of significant positive effect of government spending on capital formation on economic growth. Cooray (2009) investigated the role of the government in economic growth by extending the neoclassical production function to incorporate two dimensions of government - the size dimension (measured by government expenditure) and quality dimension (measured by governance) for a cross-section of 71 economies. The empirical results indicate that the two dimensions of government are important for economic growth. Similarly, Wu et al (2010) examined the causal relationship between government expenditure and economic growth by utilizing a panel data set which include 182 countries covering the period from 1950 to 2004, and their results provided evidence that strongly supports both Wagner’s law and the hypothesis that government spending favours economic growth regardless of how government size and economic growth are measured. By disaggregating the by income levels and the degree of corruption, their result also confirmed existence of bi-directional causality between government activities and economic growth for the different sub-samples of countries, with the exception of low income countries.

2.2 Significance of Health, Agriculture and Education in Nigeria
Health
Health services are provided by the private and public sectors. From private sector, there are non-governmental organization, private for-profit providers, community-based organization and religious and traditional care givers. Government assumes the responsibility of health service provision in public sector. The provision of health services in public sectors are at three levels namely the Primary, Secondary and Tertiary. At the primary   level, services are at the door step of communities where preventive, curative; primitive and pre-referral cares are provided. Medical personnel that provide such services are nurses, community health officers, community health extension workers (CHEWs) and environmental health officers. The available facilities at this level include health centres, dispensaries, and health.  At secondary level, there are general hospitals to provide medical, laboratory and specialized health services, namely, surgery, obstetrics, paediatrics, genecology and so on. Major health workers that are at the secondary level are doctors, nurses, midwives, laboratory scientists and pharmacists. Tertiary level of health service provision is the highest health care in the country. The facilities include specialist and teaching hospitals, and federal medical centres.
Agriculture
Agriculture played a pivotal role in the history of Nigeria’s economic development. Over the past several decades, agriculture has provided food, employment, foreign exchange and reduced poverty. It is the bedrock of Nigeria’s economy (FGN, 2004).Nigeria is endowed with a huge expanse of arable land, as well as a large, active population that can sustain a high productive agriculture. Nigeria has a great potential to become the food basket of the West African sub-region (FAO, 2003).
 Improvement in agricultural sector is a major thrust for poverty reduction. It is expected that high growth rate in agriculture will push the growth of non-faractivities as well (Gemma, 2008).
Education 
An inquiry into the fiscal operations and developments of Nigeria revealed that federal government expenditure on education is categorized under the social and community services sector. The implication is that education is an impure public good (Orubu, 1989). The importance of education is reminiscent in its role as a means of understanding, controlling, altering and redesigning of human environment (CBN, 2000). Education also improves health, productivity and access to paid employment (Anyanwu et al., 1999). Education has a link with economic development. As once remarked by Ola (1998: 14) “If you see any economy that is not doing well, find out what is spent on education”. Aboribo (1999) have all revealed that increase in national income and per capita income is a function of education and that differences among nations can better be explained by differences in the endowments of human, rather than physical capital.
The role of education in human development cannot be over emphasized. It has been described as an important tool in any human society, which makes man todevelop faster than other creatures. Education is the bedrock of all human sectors – political, medical, agricultural, security, etc. (Idogho & Imonike, 2012). Education in Nigeria is generally stratified into three sectors, which are basic, post-basic/senior secondary and tertiary education. The responsibilities for administering the education sector in Nigeria are shared among the federal, state and local governments.


CHAPTER THREE
METHODOLOGY
3.1 Source of data
 Secondary data was used in this study. The relevant data for this study have been obtained from the Central Bank of Nigeria (CBN) Annual Report and Statement of Accounts, Central Bank of Nigeria Statistical Bulletins of various years, websites and the World Bank data base. 
3.2 Simple linear regression
Regression analysis is a statistical technique for modeling and investigating the relationships between an outcome or response variable and one or more predictor or repressor variables. The end result of a regression analysis study is often to generate a model that can be used to forecast or predict future values of the response variable given specified values of the predictor variables. The simple linear regression model involves a single predictor variable and is written as
   ……………………………… (3.1)
Where Y is the response, X is the predictor variable, β0 and β1 are unknown parameters, and  is an error term. The model parameters or regression coefficients β0 and β1 have a physical interpretation as the intercept and slope of a straight line, respectively. The slope β1 measures the change in the mean of the response variable Y for a unit change in the predictor variable X. These parameters are typically unknown and must be estimated from a sample of data. The error term  accounts for deviations of the actual data from the straight line specified by the model equation. We usually think of  as a statistical error, so we define it as a random variable and will make some assumptions about its distribution. For example, we typically assume that is normally distributed with mean zero and variance, abbreviated Note that the variance is assumed constant; that is, it does not depend on the value of the predictor variable (or any other variable).
3.3 Multiple regression analysis
One independent variable such as simple linear regression model (S.L.R.M.) does not provide adequate description, since a number of key independent variable affect the respond variable (dependent variable) in important and district wages. For instant
………….…………… (3.2)
The parameters  …   in this model are often called partial regression coefficients because they convey information about the effect on y of the predictor that they multiply given that all of the other predictors in the model do not change. The regression model for cross-section data is written as
…………..……… (3.3)

Where the subscript  is used to denote each individual observation (or case) in the data set and  represents the number of observations. The regression models in Equations. (3.1) and (3.2) are linear regression models because they are linear in the unknown parameters (the), and not because they necessarily describe linear relationships between the response and the regressors.
A multiple regression is an extension of simple linear regression. It is used when we want to predict the value of two or more other variables. The variables we want to predict is dependent variables The variable we have using to predict the value of the dependent variable is called independent or predictor.
3.4 Least squares estimation in linear
We begin with the situation where the regression model is used with cross-section data. The model is given in Eq. (3.3). There are n > k observations on the response variable available, say,    .Along with each observed response, we will have an observation on each regressor or predictor variable and  denotes the  observation or level of variable.  We assume that the error term  in the model has expected value () = 0 and variance  =, and that the errors, are uncorrelated random variables. The method of least square chooses the model parameter (the ) in equation (3.3) so that the sum of the square of the errors. is minimized. The least square function is


This function is to minimized with the respect to the unknown parameters  therefore the least square estimator say  respectively must satisfy

And


By simplify equation (3.6) and (3.7) we have

3.5 Derivation of normal equations
The general linear regression model with k explanatory variables of the forms
   …………………………….…..
There are k parameters to be estimated clearly the system of normal equation will consist of k equations in which the unknowns are the parameters  and known term will be the sum of squares and sum of products in the structural equation in order to drive the k normal equations procedure we used the equation of the estimated relationship.
  ……………………….……
Where

The normal equation for a model with any number of explanatory variables may derive in mechanical way writing these normal equations
      Model with one explanatory variable 



Model with two explanatory variables
                                                                               



By comparing the normal equation of the above model we observed the following.
    The first normal equation is derived by summing the estimated form over all sample observation for instance, the estimated equation is derived by summing the estimated form over all sample observation Summing over all possible observation and using the assumption  we obtained the first normal equation.

Secondly the second normal equation is derived by multiplying the estimated form of the model by  and summing over all possible observation this gives the second equation as

Third normal equation is derived by multiplying the estimated form of the model by  and using the assumption of  as in equation  and  and summing over all possible observation

Now consider the model with k explanatory variables


The k equation of the model may obtain by multiplying the estimated form of the k variable model by  and then summing over all sample observation this gives the required  equation as


3.6 Measure of linear correlation
In the case of the above discussion it appears that we can determine the kind of correlation between two variables by direct observations of scatter diagram. In addition the scatter diagram indicates the straight of the relationship between two variables. Therefore the inspection of only the scatter diagram gives us only a rough idea about the relationship between the variables. For a quantitative measurement of the degree of correlation between the variable say X and Y we use a parameter which is called the correlation coefficient donated by Greek letter ρ. For which X and Y will be represented as   and its sample estimate is   .

When  the formula can be computed as

3.7 Hypothesis tests in multiple linear regression
This section discusses hypothesis tests on the regression coefficients in multiple linear regressions. As in the case of simple linear regression, these tests can only be carried out if it can be assumed that the random error terms, ε, are normally and independently distributed with a mean of zero and variance of σ2. Three types of hypothesis tests can be carried out for multiple linear regression models:
1. Test for significance of regression: This test checks the significance of the whole regression model.
2.   Test: This test checks the significance of individual regression coefficients.
3.  Test: This test can be used to simultaneously check the significance of a number of regression coefficients. It can also be used to test individual coefficients.

3.8 Estimating
We wave introduced the concept of estimating of regression coefficient  in equation (3.8) above it is also necessary to estimate the variance of the model error .the estimator of this parameter involve the sum of square residual we can show that

So that estimator of  is the square error

3.9 Coefficient of multiple determination 
The coefficient of multiple determinations is similar to the coefficient of determination used in the case of simple linear regression. It is defined as:

Indicates the amount of total variability explained by the regression model. The positive square root of is called the multiple correlation coefficient and measures the linear association between  and the predictor variables,  .
 The value of   increases as more terms are added to the model, even if the new term does not contribute significantly to the model.  An increase in the value of    cannot be taken as a sign to conclude that the new model is superior to the older model. A better statistic to use is the adjusted statistic defined as follows:

3.1.2 Tests for the Existence of Multicollinearity Problem on the Data Sets
You will recall that one of the assumptions of regression analysis is that the explanatory variables are independent of each other, that is, two or more explanatory variables do not tend to move together in the same pattern. When this assumption fails, we say there is multicollinearity among the independent variables.

3.1.1 Discussion of the result
Table    4.2    Descriptive Statistics


REC.EXP
HEALTH
AGRICULTURE
EDUCATION


N
13
13
13
13








Mean
1981.2054
202.8354
72.4723
302.9023

Std. Error of Mean
208.57553
21.86055
11.15225
34.66825

Median
2386.0200
235.8600
75.8100
306.3000

Std. Deviation
752.02977
78.81935
40.21000
124.99815

Variance
565548.768
6212.490
1616.844
15624.537










The table 4.2 shows that the overall average, Std. Error of Mean, median, Std Deviation, and variability (standard deviation) of each variable).







Table 3.1 Results of Computation and Significance Test of Individual Regression Parameters.


Model
Unstandardized Coefficients

 

Collinearity Statistics


B
Std. Error


VIF

1
(Constant)
35.370
197.994
0.179
0.862

18.023
1.097
17.738


HEALTH
8.227
3.589
2.292
0.048



AGRICULTURE
2.276
1.736
1.311
0.222



EDUCATION
0.371
2.245
0.165
0.873



Here, we present estimates of regression parameters using ordinary least squares technique. Table 3.1 presents summary of results for the estimation and test of significance of regression parameters. It is evident from the results that only Health (is significant since its value is less than significance level (0.05).                         The estimate regression equation can therefore be fitted as follows.



Since and are very large It is reasonable to conclude that there is multicollinearity problem in the data sets.







Table 3.2 Analysis of Variance (ANOVA) Results Table for Overall Test of Significance



Source


DF


SS


MS


F


P



Regression

Residual Error

Total


3

9

12


6307126

479460

6786585


   2102375

53273



39.46




0.000







The table above reveals that the full regression model is sufficient since its value  is less than significance level (0.05). The implication of this is that there is linear relationship between the  Recurrent Expenditure and its component simultaneously, As a result of this; we can reliably conclude that the estimates obtained in table 3.1 can be used to model the relationship between  Recurrent Expenditure and its component under consideration. 















Variable
     Total Recurrent Expenditure
Health
Agriculture
Education



     Total Recurrent Expenditure
1





0.929


0.906

Health
0.957
0.000

1






Agriculture
0.385
0.193
0.287
0.341
1




Education
0.931
0.000

0.971
0.000

0.260
0.390
1



Table 3.3 Results of Computation of Simple Correlation Coefficients between the Recurrent Expenditure and its Component as well as Coefficient of Determination



Interpretation of the Coefficients of Correlation and Determination
So far, there is a clear indication that only Health and Education can be statistically modeled with Total Recurrent Expenditure. Therefore, these further confirm the allegation such that only Health and Education has approximately 96% and 93% respectively positive linear relationship with the Total Recurrent Expenditure. The implication of this is that the rate at which increase or decrease in Health and Education leads to corresponding increase or decrease in  Recurrent Expenditure is approximately 96% and 93% respectively. Coefficient of determination is 93% indicating that only 7% variation cannot be explained by Health and Education. The implication of this is that 93% of all financial activities depend on Health and Education.








Table 3.4

Stepwise Regression: Total Recurrent versus Health, Agriculture, Education

Backward elimination.  Alpha-to-Remove: 0.1


Response is  Recurrent Expenditure on 3 predictors, with N = 13


Step               1       2       3
Constant       35.37   32.85  129.16

Health          8.23    8.80    9.13
T-Value         2.29   10.50   10.94
P-Value        0.048   0.000   0.000

Agriculture      2.3     2.3
T-Value         1.31    1.37
P-Value        0.222   0.200

Education        0.4
T-Value         0.17
P-Value        0.873

S                231     219     228
R-Sq           92.94   92.91   91.58
R-Sq(adj)      90.58   91.50   90.82
Mallows Cp       4.0     2.0     1.7
PRESS        1193805  852894  754826
R-Sq(pred)     82.41   87.43   88.88
















Table 3.5

Best Subsets Regression: Total Recurr versus Health, Agriculture, ...

Response is Total Recurrent Expenditure

                                          A
                                          g
                                          r E
                                          i d
                                          c u
                                        H u c
                                        e l a
                                        a t t
                                        l u i
                       Mallows          t r o
Vars  R-Sq  R-Sq(adj)       Cp       S  h e n
   1  91.6       90.8      1.7  227.89  X
   1  86.6       85.4      8.1  287.39      X
   2  92.9       91.5      2.0  219.30  X X
   2  91.6       89.9      3.7  238.97  X   X
   3  92.9       90.6      4.0  230.81  X X X



 The suitable models connecting  Recurrent Expenditure with its components is obtained as follows:

y(Rec.Exp)  = 35.37+8.23Health+2.3Agriculture+0.4Education
y(Rec.Exp)  = 32.85+8.80Health+2.3Agriculture
y(Rec.Exp)  = 129.16+19.13Health













Figure 3.1


Figure 3.2




Figure 3.3


From figures 3.1 and 3.3, it is evident that linear relationships exist between Health and  Recurrent Expenditure as well as between Education and  Recurrent Expenditure whereas figure 3.2 pictures out a non linear relationship between Agriculture and Total Recurrent Expenditure. The preview of the type of relationship(s) has already been identified through these various scatter plots.

















Figure 3.4




Figure 3.4, it is inferred that normality assumption is not violated. Since the points clustered around a straight line, the assumption of normal p-p plot is meeting 
















CHAPTER FOUR
SUMMARY CONCLUSION AND RECOMMENDATION
4.1 Summary
 In chapter one; the study provide a general introduction, Aim and objectives, significance of the study, Scope of the study, Research hypotheses, as well as Definition of terms (Health; Agriculture; Education and recurrent Expenditure).
Chapter two contains related review of the study.
In Chapter three discussions were made as regards to Source of data, Source of data, Simple and multiple linear regression analysis presentation and interpretation of data.
Chapter four contains summary conclusion and recommendation references. 
4.2 Conclusion 
The project investigate the relationship between government expenditure (Health, Agriculture, Education) on Recurrent Expenditure in Nigeria from 2005-2017. A multiple regression analysis was done. The variables were tested using analysis of variance (ANOVA) and correlation was also performed. The result shows that in correlation test, on Health, Agriculture and Education were found to be positively related to total recurrent expenditure moreover the result in analysis of variance shows the overall significance and indicates that there is linear relationship between the expenditures. And the best regression model connecting Health, Agriculture and Education on Recurrent Expenditure in Nigeria is
y(Total Recurrent Expenditure)  = 129.16+19.13Health

4.3 Recommendation
On the basis of the result obtained the following recommendation will be essential:-
Firstly, government should ensure that Total Government Expenditure is properly managed in a manner that it will raise the nation’s production capacity and accelerate economic growth.
Secondly, Nigerian budget allocation to agriculture is less than the 10 percent  NEPAD target of national budget and should therefore, be increased by act of legislation so that agriculture projects will be effectively implemented.
Thirdly, government should encourage the Education sectors through increased funding, as well as ensuring that the resources are properly managed and used for the development of education services.
 Fourthly the government should provide more found for agricultural universities in the country to carry out more research on all aspect of agricultural output such as livestock, crops, fishing and forestry, crops presentation.
 Lastly, government should increase its funding of anti-graft or anti-corruption agencies like the Economic and Financial Crime Commission (EFCC), and the Independent Corrupt Practices Commission (ICPC) in order to arrest and penalize those who divert and embezzle public funds.




















REFERENCES

Abdullah H. A (2000), “The Relationship between Government Expenditure and                     Economic growth in Saudi Arabia”, Journal of Administrative Science, 12(2):173–191.

Abu & Abdullahi (2010), “Government Expenditure and Economic growth in Nigeria 1970 2008: A Disaggregated Analysis”, Business and Economics Journal, 4: 1-11.

Adeniyi, O. and Abiodun, N.(2011). Health Expenditure and Nigerian Economic Growth. European Journal of Economics, Finance and Administrative Sciences.
Ajayi, K. & Adeniji, A. (2009). Access to university education in Nigeria. In B.G. Nworgu & E.I. Eke. (Eds.), Access, quality and cost in Nigerian education (Pp.35- 60). Published Proceeding of the 23rd Annual Congress of the Nigerian Academy of Education.

Akpan N. I (2005), “Government Expenditure and Economic growth in Nigeria: A Disaggregated approach”, CBN Economic and Financial review, 43(1).

Baro R, Sala-I-Martinx (1992), “Public Finance in models of Economic growth”, Review of Economic studies, 59: 645–661.
 Barro R, (1990). Government Spending in a Simple Model of Endogenous Growth. Journal of Political Economy, 98(5):103-125.
Central Bank of Nigeria (CBN), (2006).  Annual Report and Statement of Accounts. Abuja: CBN.
Central Bank of Nigeria (CBN),  (2006).  Statistical Bulletin, Vol. 17. Abuja: CBN.
Central Bank of Nigeria (CBN), (2007).  Finance, Investment and Growth in Nigeria. www.cenbank.org. (accessed Nov. 12 2008).
Central Bank of Nigeria (2010): Statistical Bulletin Central Bank of Nigeria. Vol.21, December. Chude, N.P and Chude, N.I. (2013). Impact of Government Expenditure On Economic Growth In Nigeria. International Journal of Business and Management Review Vol.1, No.4, pp. 64-71, December.
Cooray A (2009), “Government Expenditure, Governance and Economic growth: Comparative Economic studies”, 51(3): 401–418.
Dickey, D., Fuller, W., (1979). "Distribution of the Estimators for Autoregressive Time Series with a Unit Root", Journal of the American Statistical Association, Vol. 74, pp. 427-31. 
Easterly W & Rebelo S (1993), “Fiscal Policy and Economic growth: An Empirical Investigation”, Journal of monetary Economics, 32: 417–458
Fajingbesi A. A & Odusola A. F (1999), “Public Expenditure and growth”, A paper presented at a training programme on Fiscal Policy Planning Management in Nigeria organized by NCEMA, Ibadan, Oyo state, 137–179.
 Folster, S & Henrekson M, (2001), Growth Effects of Government Expenditure and Taxation in rich countries. European Economic Review, 45(8): 1501–1520.
Grier, K. and G. Tullock, (1989). An empirical analysis of cross-national economic growth, 1951-1980, Journal of Monetary Economics 24, 259-276
 Ihugba, O.A., Nwosu, C. and Njoku, A.C. (2013).  An Assessment of Nigeria Expenditure on the Agricultural Sector: It’s Relationship with Agriculture Output (1980 - 2011). Journal of economics and international finance Vol. 5(5), pp. 177-186, August. http://www.academicjournals.org/JEIF.
 Jiranyakul, K. and Brahmasrene, T. (2007). The relationship between government expenditures and economic growth in Thailand. Journal of Economics and Economic Education Research. 
Kormendi, R.C., and Megurie, P.G., (1985). Macroeconomic determinants of growth: cross-country evidence, Journal of Monetary Economics 16, 141-63
Company. Ogiogio, G. O (1995), Government Expenditure and Economic growth in Nigeria. Journal of Economic Management, 2(1).