Contents
What if the coefficient is not statistically significant?
If the t-test for a regression coefficient is not statistically significant, it is not appropriate to interpret the coefficient. Even when there is an exact linear dependence of one variable on two others, the interpretation of coefficients is not as simple as for a slope with one dependent variable.
How do you know which predictor is most important?
Generally variable with highest correlation is a good predictor. You can also compare coefficients to select the best predictor (Make sure you have normalized the data before you perform regression and you take absolute value of coefficients) You can also look change in R-squared value.
When to remove a non-significant predictor from a model?
When I remove this the AIC value goes up indicating the model without the significant variable is a worse fit. Should I remove the non-significant predictor or should I leave it in as it is a better model? Leave it in.
Is it bad to select predictors based on significance alone?
Consequently, material here and elsewhere about non-significant terms in a linear model are relevant, as are the well known critics of stepwise regression, even if ANCOVA is not explicitly mentioned. Generally speaking, it’s a bad idea to select predictors based on significance alone.
What are non signficant coefficients of factor levels?
After seeking clarification about linear model coefficients over here I have a follow up question concerning non-signficant (high p value) for coefficients of factor levels.
When to not include coefficient term in linear model?
Example: If my linear model includes a factor with 10 levels, and only 3 of those levels have significant p values associated with them, when using the model to predict Y can I choose to not include the coefficient term if the subject falls in one of the non-signficant level?