Application of hypothesis testing in estimating regression models

Abstract

Hypothesis testing is a typical technique used in data analysis to ascertain the accuracy of an estimated regression model obtained from the data. Therefore, this method is closely related to the accuracy of the conclusions drawn. There are many studies that have used hypothesis testing to draw conclusions, however, there is still a lack of generalizations about the accuracy of the estimated model. Therefore, this paper will shed light on how the accuracy of estimated regression models can be analyzed through F-tests as well as specification tests. The F-test confirms the significance of the overall fitness of the model, yields whether there are any omitted or irrelevant variables. Finally, the specification test confirms whether each variable should be included in the model. The accuracy of the estimated regression model can be interpreted after completing the specification test for each individual variable to obtain a final model that is statistically significant using the F-test. This work highlights the importance of hypothesis testing in estimating various regression models.