The Impact Of Interest Rate On Domestic Investment

METHODOLOGY

This study makes use of econometric procedure in estimating the impact of interest rate on economic growth in Nigeria. The Ordinary Least Square (OLS) technique is employed in obtaining the numerical estimates of the coefficients in different equations. The OLS method is chosen because it possesses some optimal properties: its computational procedure is fairly simple and it is also an essential component of most other estimation techniques. Secondly, this technique has been adopted by many other researchers and has yielded optimal results. Thirdly, the OLS technique possesses the feature of simplicity as it is not complex in computation. Finally, the data requirement of this technique is not excessive as compared with other techniques and methods.

3.1 Theoretical Framework

The theoretical basis of this research is the Keynesian theory of interest rate. This theory assumes a closed economy and a perfect competitive market with fairly price- interest aggregate supply function. The economy is also assumed not to exist at employment equilibrium and also that it works only in the short run because as Keynes aptly puts it ‘’ In the long run, we also will be dead’’. The Keynesian theory is rooted on one notion of price rigidity and possibility of an economy setting at a less than full employment level of output, income and employment.

3.2 Model Specification

In this research, interest rate will be the independent variable with inflation rate and money supply as control variables while domestic investment will be the dependent variable. The model is thus:

By Definition:

DI = Domestic Investment

INTR = Interest Rate

INF =     Inflation Rate

MS = Money Supply

µ = Stochastic Error Term

1. 1. Method of Evaluation

3.3.1 Preliminary Tests

3.3.2 Economic Criterion Test (A priori Test)

The a priori test of the analysis will be based on the regression coefficient based on the coefficient of the algebraic signs of the parameters. It is a test that will be based on evaluating the conformity of the relationship between the variables on economic theory.

3.3.3 Statistical Test of Significance

3.3.3.1 Test for Goodness of Fit

This test involves the test of the goodness of fit. To evaluate the working hypothesis of this study. R2 the co-efficient multiple of determination is used to test the explanatory power of the variable. R2 lies between zero and one (0 < r2 < 1). The closer R2 is to 1 the greater the proportion of the variation in the dependent variables attributed to the independent variables.

3.3.3.2 T-Test of Significance    

To test for the statistical significance of individual regression co-efficient, t-statistic is used. A two-tailed test will be conducted at 5% level of significance and n-k degree of freedom; where, n = sample size and k = number of parameters in the model. The null hypothesis Ho will be tested against the alternative hypothesis H1.

Decision Rule (T-Test)

If t0.025 < t* Ho will be rejected and the H1 accepted. Otherwise, the alternative hypothesis H1 will be rejected and the null hypothesis Ho be accepted.

 3.3.3.3 f-test of Significance

To Test the statistical significance of the entire regression, the f-ratio is used. The test will be conducted at 5% level of significance and V1/V2 degree of freedom (df); where, V1 = df for the numerator and V2 = df for the denominator.   

Decision Rule (F-Test)

If f* > (f0.05), we say the regression is statistically significance but if otherwise, it implies that it is statistically insignificant

Note: t*       =        computed t – value

                   t0.025 = tabulated t – value

                   f*       =       Computed f-value

                    f0.05    =       tabulated f – value

3.3.4 Econometrics Test of Significance

3.3.4.1 Autocorrelation Test

To evaluate the reliability of the expected numerical estimates, the Durbin – Watson (D-W) statistics at 5% will be used to test for the presence of autocorrelation problem. The region of autocorrelation remains:

du < d* < (4-du)

Where:

du = Upper Durbin – Watson

d* = Computed Durbin-Watson

Decision Rule (Autocorrelation Test)

If the computed value of Durbin-Watson lies within the region, it means there is no presence of autocorrelation problem. But if the Durbin-Watson computed value lies outside the regions there is the presence of autocorrelation problem and a remedial measure like the use of first difference equation will be adopted. 

3.3.11 Unit Root/Stationary Test

This is will be used to test whether a variable’s mean value and variance varies over time. It is necessary in time series variables in order to avoid the problem of spurious regression. The Augmented Dickey Fuller (ADF) test will be used for the analysis.

3.3.1.2 Co-integration test

This will be used to test if there exists a long-run relationship between the variables under investigation. The Johansen or Engel-Granger methodology will be used.

3.3.1.3 Error Correction Model (ECM)

If cointegration exists among the variables, the error correction model will be estimated. The ECM is estimated to estimate the short-run dynamics and the speed at which the short run disequilibrium corrects to achieve a long-run relationship.

 3.4 Data Required and Sources

The data required for this research are time series secondary data on interest rate, inflation rate, money supply and domestic investment. Data will be extracted from the Central Bank of Nigeria (CBN) statistical bulletins and National Bureau of Statistics (NBS).