Goodness-of-fit: Desirable Properties of a Regression Model

In interpreting the quality of a regression model, the following aspects are taken into consideration.

High adjusted R2: R2 is proportion of variation in the dependent variable explained by the independent variables. Radj2 adjusts for number of explanatory terms in a model relative to the number of data points (i.e., df associated with the sums of squares).

Significant F-statistic: The null hypothesis that all coefficients (b1, b2) are zero can be conclusively rejected.

Significant t-statistics: The individual coefficients are statistically significant.

Residuals are random, normally distributed, with mean 0, and their variance does not vary significantly, over the values of the independent variables.


Previous     Next

Use the Search Bar to find content on MarketingMind.