Dear Phil Chan, if we have more continuous variable, can we drop the intercept? or this way ( k dummies ) just use when all independent variable are category?
+Khamphonemisay PHOMMATHAT With k-1 dummies, the coeff on a dummy tells you the mean change in the DV of that group compared to the reference group. If you use k dummies the coeff now tells you the mean value of the DV for that group.
+Phil Chan (statisticsmentorcom) Thank you , that's a better way for interpretation with K dummies. It is easy to understand the effect of each dummy contribute to dependent variable because I though that the sign ( negative or positive ) of Beta is important , not the intercept ( beta zero ).