Regression predictions are for the mean of the dependent variables. However, the values that drew my attention in this regression model are the R-Squared values and how they were adjusted. Additionally, the value of the R-squared was good but not perfect since higher R-squared values represent predictions that are mode precise. Additionally, the standard error of regression of the regression was not given thus it’s difficult to assess the precision of the model. In a standard regression model, independent variables should be entered one at a time, while the predictor variable should be excluded if it can be perfectly predicted from one or other independent variables. This is because the adjusted R-squared, any other variable that has a greater t-value than the absolute value of 1 will increase the R-squared.
I am not satisfied with the presented regression mode predictor since it is unclear whether the residual plots were checked. This is important because residual plots can be used to avoid making a biased model and should have helped to make adjustments. Besides, this regression model cannot tell which equation is best. The business analyst could have also excluded the f-ratios and the negative t-values. Additionally, the t-values that are near 1or -1 are not statistically significant. It would have also been essential for the business analyst to fit the model by increasing the adjusted R-squared so that the insignificant t-values can be included but do increase the adjusted R-squared. Finally, based on the predictions made above, it implies that the business analyst did not predict new observations about the new markets that the company intends to expand to. The analysis has also over-fitted the model.
Delegate your assignment to our experts and they will do the rest.
