ABSTRACT

The ma30r objective of the study is to construct an appropriate time series model which could be used to predict the production of agricultural output in Nigeria. The data was collected from the central bank of Nigeria (CBN) Statistical Bulletin for the period of 1982 — 2012 was used for the study. The Box-Jenkins iterative method was used. The plot of the original data of agricultural output and its associated autocorrelation and partial autocorcelation plot showed that the data is non-stationary that is to say; there is presence of trend and seasonal variations in the data. A differencing of order 1 was applied to remove the trend, the plot of autocorrelation function and partial autocorrelation function shows a repeated pattern at every lag 4 and the seasonal effect remaining. A differencing of order 8 (for quarterly) vvas applied to remove the nal variation. The plot of autocorrelation function of order 8 show ad an alternating pattern and the partial autocorrelation function plots becomes zero after lag 2. Finally, SARIMA (1,1,1) x (2,2,2)4 model was fitted to the agricultural output data for the period under consideration.

TABLE OF CONTENTS

Title Page

Certification ii

Declaration iii

Dedication W

Acknowledgment V

Table of Contents Vi

abstract

CHAPTER ONE

1.0 tctcoduct(on 1

1.1 Background of the Study 2

1.1.1 Agricultural Crops in Nigeria are Grouped into the following 4

1. .2 Stateect of the Prob1ecr 4

1.3 Objectives of the Study 5

1.4 Significance of the Study 6

1.5 Scope of the Study 6

1.6 Plan of the Study 7

CHAPTER TWO

2.0 Uteratuce Review S

2.1 Agricultural Policies During The Military Era 10

2.1.1 Agricultural Policies under Current Democratic Dispensation 13

2.2 Major ChaUenges of Agricultural Production 14

• CHAPTER THREE

3.Q Data Collection and Method of Analysis 16

3.1 Data Source 16

3.2 Definition of Terms 16

3.3 Method of Analysis 17

3.3.1 Autocorrelation Function (ACF) 17

3.3.2 Properties of Autocorrelation Functions 18

3.3.3 Partial Autocorretation Function 18

3.4 Models for Stationary Time Series 19

3.4.1 Autoregressive Process, AR(P) 19

3.4.2 Moving Average Process MA(q) 20

3.4.3 Autoregressive Moving Average Process (ARMA) p,q 21

3.4.4 Autoregressive Integrated Moving Average (ARIMA) (p,d,q) 22

3.4.5 Seasonal ARIMA Model (SARIMA) 22

3.4.6 Building of Model 23

3.4.7 Model Identification 23

3.4.8 Estimation of Parameter 23

3.4.9 Diagnostic Checking 24

CHAPTER FOUR

4.Q Results and Discussion 26

4.1 Time Plot of the Original Series 26

4.2 Estimation of Parameters 32

. .3 Residual natysis 33

CHAPTER FIVE

S.Q Sunrnary, Conclusion and Recornniendations 34

5.1 Summary 34

5.2 Conclusion 35

5.3 Recomcnendation 35

References

Appendix