Acknowledgement

Ordinary least square assumptions

• 
  The parameters of the model regression are linear
  

• 
  As seen in the table above, each of the two subgroups comprises 98 observations, and each is regressed using the five independent variables.
  

• 
  The test of variance inflation factor, also known as the VIF test, was also performed using Stata software, and the results revealed that there was no multi-collinearity because the average value of the indexes was less than 10, indicating that there was no perfect correlation between the independent variables. (see the graph below)
  

• 
  Additionally, the study includes assessments of the constant variance var(ui|Xi), which were discovered using the Breusch-Pagan test, with the results indicating that the measurement process is free of heteroscedasticity because the value of p is less than 0.05. (see the graph below)
  

• 
  Furthermore, the results of the DURBIN WATSON TEST, which was performed in Stata, suggested a value of 2.6. Autocorrelation is present when the value is near to two, indicating that there may be some breaches of this assumption. (see the graph below)
  

• 
  In addition, Stata performed a normality assumption test using the Kernel density model; as seen in the image below, the residual density looks to be close to normality, indicating that there are no violations of this assumption. (see the graph below)
  

• 
  STATA was also used to run the test for omitted variable measurement. The results yielded a P=1.54 value, showing the absense of missing factors. (As shown in the graph below.)
  

• 
  The results of the correlation and covariance tests were as follows: