HYPOTHESIS TESTING

(Statistics) The condition whereby the Null Hypothesis, which argues against the validity of the model's structure (Specification) is tested using various statistical criteria, e.g., Coefficient of Determination, or R2, t-Statistic, F-Statistic, etc., against the Alternative Hypothesis that there exists a significant relationship or correlation between the dependent variable, or variable to be explained, and the independent, or explanatory variable(s). In hypothesis testing, a rule for acceptance and rejection must be chosen, e.g., 5 percent level of significance, that is, there exists a 5 percent chance that in rejecting the null hypothesis, which states that the disturbances in the dependent variable are purely random, we will be wrong. Inversely, there is a 95 percent chance that by rejecting the null hypothesis and accepting the alternative hypothesis, we will be right.