I am running a regression on Tax payment and investment. Tax payment is numerical and investment is categorical like, number of projects carried out. When I run the regression the a unstandardised beta coefficient is very very small, close to zero, but standardised coefficient is about 0.348. The p value is less than 0.05 and therefore statistically significant. Appreciate your quick response as I am running out of time in completing the thesis.

Best regards Sudhuna


1 Answer 1

  • You need to scale the data only for certain kinds of models, for numerical reasons. For example, when using regularization. If you are using a regression model without regularization, or random effects, you don't need it.
  • In some cases, you can use scaling if it makes the parameters easier to interpret.
  • Scaling does not affect your results (for standard regression, as said above), predictions, and $p$-values would stay the same. The only thing that would change is the scale of the parameters because they reflect the scale of the features, for example, if you used a feature in minutes vs hours, in the second case the parameter for it would be 60 times smaller.

See also When conducting multiple regression, when should you center your predictor variables & when should you standardize them?


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.