CHAPTER THREE

METHODOLOGY

3.1 Introduction

This chapter outlines the methodological framework used to investigate the impact of Deposit

Money Banks (DMBs) on the Agricultural Sector in Nigeria. It encompasses data sources,

sampling techniques, variable measurements, model specification, and the estimation technique

employed.

3.2 Sources and Types of Data

The study utilizes annual time series data spanning from 1980 to 2022. Data on the dependent

variable, agricultural output (AO), was sourced from the National Bureau of Statistics (NBS,

2023) agricultural reports. However, the data for independent variables, including deposit

money bank credit (DMC) and deposit money bank interests (DMI), was obtained from the

Central Bank of Nigeria (CBN, 2023) statistical bulletin. Finally, data on government

expenditure was also obtained from the National Bureau of Statistics (NBS, 2023) database.

The selection of these specific data sources was guided by the principle of credibility and

accessibility. Both the National Bureau of Statistics and the Central Bank of Nigeria are

renowned governmental institutions in Nigeria, recognized for their meticulous data collection

practices and adherence to rigorous statistical methodologies. Furthermore, utilizing official

publications and databases ensured the data's accuracy and consistency.

1
3.3 Sample Size and Sampling Technique

This research employs all available data for the period 1980-2022, resulting in a sample size of

43 observations. This approach ensures the capture of the maximum information available and

avoids potential biases associated with smaller sample sizes (Gujarati & Porter, 2009). By

employing a comprehensive data collection strategy and a robust sample size, this research

endeavored to establish a solid foundation for a rigorous analysis of the factors influencing

3.4 Variables Measurement

Table 3.1 Variables Measurement and Definition

S/N Variable name Level Type Measurement Data source

1. Agricultural AO Dependent Measured as the real National
Output value of agricultural Bureau of
production at constant Statistics
prices (e.g., base year (NBS, 2023)
2010) database.
2. Deposit Money DMC Independent Measured as the total Central Bank
Banks Credit credit extended by of Nigeria
deposit money banks to (CBN, 2023)
the agricultural sector statistical
bulletin
3. Deposit Money DMI Independent Measured as the Central Bank
Banks Interests average interest rate of Nigeria
charged by deposit (CBN, 2023)
money banks on loans statistical
to the agricultural bulletin
sector
4. Government GE Independent Measured as the total National
Expenditure government Bureau of
expenditure on Statistics
agriculture (NBS, 2023)
data base
Source: Authors computation, 2024

2
Table 3.1 above presents and explains the variables employed in this study by specifying the

variable's name, description, and measurement. However, to justify the choice of these variables

as they relate to the study, the following discussion is immensely important.

Agricultural Output (AO): This variable is chosen as the dependent variable in all three

research questions outlined in chapter one of this study. The variable directly addresses the

study's focus on the impact of various factors on agricultural production in Nigeria. Agricultural

output serves as a quantifiable measure of the sector's performance, allowing for the assessment

of various interventions' impact on production levels.

Deposit Money Banks Credit (DMC): This variable is included in research question (i) to

examine its potential influence on agricultural output. This selection is supported by numerous

empirical studies demonstrating a positive relationship between financial inclusion and

agricultural development. Studies by Magaji et al. (2023), Asukwo et al. (2020), Chris et al.

(2016), Udoka et al. (2016), Ogbuabor & Nwosu (2017), Oyelade (2019), Ebere et al. (2021),

Onuegbu et al. (2022), Florence & Nathan (2020), Okafor (2020), George-Anokwuru (2018),

Olu et al. (2023), and Azike et al. (2020) consistently report a positive and significant impact of

bank credit on agricultural output in various contexts. Increased access to credit through bank

loans empowers farmers to invest in crucial inputs, equipment, and technology, potentially

leading to enhanced productivity and, consequently, higher output.

Deposit Money Banks Interest Rate (DMI): This variable is included in research question (ii)

to analyze its impact on agricultural output. This selection is grounded in the understanding that

higher interest rates can discourage farmers from seeking loans, thereby limiting their investment

capacity and potentially hindering output growth. This rationale is supported by research on

3
access to finance and agricultural productivity, as evidenced in studies by Chris et al. (2016),

Oyelade (2019), and Azike et al. (2020). These studies highlight the negative, albeit sometimes

insignificant, impact of interest rates on agricultural output, suggesting that higher borrowing

costs can act as a barrier to investment and production growth.

Government Expenditure on Agriculture (GE): This variable is included in research question

(iii) to assess the influence of government policies on Agricultural output. This choice is justified

by the potential impact of government spending on various initiatives, such as infrastructure

development, extension services, and subsidies, on agricultural production levels. The relevance

of this variable is further supported by existing literature on government policies and agricultural

development, as demonstrated in studies by Chris et al. (2016), Ebere et al. (2021), Adofu

(2012), and Iganiga & Unemhilin (2011). These studies consistently report a positive relationship

between government expenditure on agriculture and output, suggesting that targeted investments

can stimulate production growth and enhance the sector's performance.

3.5 Model Specification

The following econometric model is specified to analyze the impact of Deposit Money Banks on

the Agricultural Sector in Nigeria:

AO = α + β₁DMC + β₂DMI + β₃GE + ε…………………………………………………… (3.1)

Where:

AO = Agricultural Output and it also stands as a proxy variable for the agricultural sector in
Nigeria

DMC = Deposit Money Banks Credit

DMI = Deposit Money Banks Interests

4
GE = Government Expenditure

α = Constant term

β₁ = Coefficient of Deposit Money Banks Credit

β₂ = Coefficient of Deposit Money Banks Interests

β₃ = Coefficient of Government Expenditure

ε = Error term

This model assumes a linear relationship between the dependent and independent variables.

3.6 Estimation Technique

The Ordinary Least Squares (OLS) method will be employed to estimate the parameters of the

specified model. The choice of Ordinary Least Squares (OLS) as the estimation technique for

this analysis is based on several key factors. Firstly, the specified model posits a linear

relationship between the dependent and independent variables. This aligns perfectly with the

core assumption of OLS, making it the most suitable method for capturing the hypothesized

relationships (Greene, 2008). Secondly, OLS estimates offer a significant advantage in terms of

interpretability. The coefficients obtained through OLS directly represent the average change in

the dependent variable associated with a one-unit change in the corresponding independent

variable, holding all other factors constant. This straightforward interpretation facilitates clear

economic analysis and allows for the formulation of well-defined policy recommendations based

on the estimated effects (Wooldridge, 2010).

Finally, under specific assumptions, OLS boasts desirable statistical properties. These include

unbiasedness, consistency, and efficiency. Unbiasedness ensures that the expected value of the

estimated coefficients accurately reflects the true population parameters, leading to reliable

5
estimates on average. Consistency guarantees that as the sample size increases, the OLS

estimates converge toward the true population parameters, providing increasingly accurate

results with larger datasets (Stock & Watson, 2018). Lastly, efficiency implies that among all

linear unbiased estimators, OLS minimizes the variance of the estimates, resulting in the most

precise estimates achievable under the given assumptions.

However, before proceeding with estimation, several diagnostic tests will be conducted to ensure

the validity of the OLS results.

3.6.1 Unit Root Test:

Unit root tests, such as the Augmented Dickey-Fuller (ADF) or Phillips-Perron (PP) tests, will be

conducted to determine the stationarity of the variables. Non-stationary variables can lead to

spurious regressions. According to Oyeni, (2012), "a spurious regression occurs when a non-

stationary time series variable is regressed on one or more non-stationary time series variable(s)

which makes such results invalid and therefore cannot be used for forecasting". Therefore, this

study will adopt the Augmented Dickey fuller (ADF) unit root test to verify the unit property of

the series. All the variables in the models will be subjected to the ADF test. The Augmented

Dickey fuller (ADF) unit root test is specified as follows:

yt 01 yt1iyt1t …………………………………………………….……. (2)

Where ∆ yt is the variation in у at period t, α0 represents constant, yt-1 is the past value of у, α1

is the estimated lag coefficients and µt is the error terms. The existence of unit root can be

determined through the following hypotheses: Null hypothesis is defined as H 0: = 0 (series have

unit root). The alternative hypothesis is given as H1: < 0 (series variables are stationary or series

does not have unit roots).

6
3.6.2 Test for Serial Correlation:

The presence of serial correlation in the residuals can lead to inefficient and biased estimates.

The Breusch-Godfrey LM test will be employed to detect and correct for serial correlation, if

necessary (Breusch, 1979).

3.6.4 Normality Test:

The Jarque-Bera test will be used to assess the normality of the error term. While not strictly

necessary for OLS estimation, normality can improve the efficiency of the estimates (Jarque &

Bera, 1980).

3.6.5 Stability Test:

The CUSUM and CUSUMQ tests will be conducted to assess the stability of the estimated

coefficients over the sample period. This helps ensure the reliability of the model's results

(Cumby & Obstfeld, 1994).

By employing these diagnostic tests and addressing any potential issues, the research aims to

ensure the robustness and reliability of the estimated results and draw meaningful conclusions

about the impact of DMBs on agricultural output in Nigeria.