S z research article

 
RESULTS 
 
Data Analysis and Interpretation: This study employs 
STATA 12 package to carry out Descriptive Statistics, 
Correlation and Regression Analysis while testing for auto-
correlation as follows; 
Table 1. Durbin Watson Test 
 
Variable 
Durbin-Watson d-statistic 
ROA 
1.087453 
 
Durbin Watson Test: The study employed Durbin Watson 
test to test the presence of autocorrelation. From table (1) 
below the value of 1.087453 for the Durbin Watson statistic 
indicates that the null hypothesis of no autocorrelation is not 
rejected. This implies that the residuals are not serially 
correlated. 
 
Descriptive Statistics: Descriptive statistics gives initial 
indication of variables that can be used in regression analysis 
giving several summarized statistics on a variable. In this study 
descriptive statistics consist of five variables that are 
independent variables (GDP Rates, INFR, INTR, EXCR 
Tshs/Usd and Government Debts) and dependent variable 
(ROA) as it is shown in the table (2) below shows the 
characteristics of the variables used by revealing the statistical 
mean, standard deviation, minimum, and maximum values. 
The findings reveal that the average of bank performance over 
the period from 2011 to 2019 is 2.84356%. The results indicate 
that bank performance (ROA), economic growth (GDP), 
inflation interest rate and Government debts do not deviate 
much from the mean (the variables have smaller standard 
deviation) thus the more accurate are the future predictions. 
 
Regression Analysis: Table (3) reports regression analysis 
about the impact of macroeconomic variables on bank 
performance, the study employed multiple regression analysis 
using Pooled Ordinary Least Square Regression Model. The 
findings reveal that economic growth (GDP) has an 
insignificant positive relationship between bank performances 
(ROA) at 10 % level of significance. Therefore does not reject 
the null hypothesis. This is consistent with the findings of 
Kanwal, Nadeem, (2013). It is found that, inflation has a 
negative and insignificant effect on bank performance at 10 % 
level of significance. Thus, the result does not reject null 
hypothesis that inflation has inverse relation with banks’ 
performance. This finding is consistent with the empirical 
study by Sanusi, M., Zulaikha, S. (2019). Furthermore, the 
results indicate that exchange rate has an insignificant negative 
effect on bank performance at 10 % level of significance. The 
study therefore does not reject the null hypothesis that 
exchange rate affect the banks’ performance negatively. The 
Government Debts has statistically significant with negative 
effect on bank performance at 10 % level of significance. This 
finding is consistent with the empirical study by Pegkas, 
(2018), Eze, et al. (2019),