Changes

Jump to navigation Jump to search
The crucial information is the magnitude of effect of coefficients:
*Log-level: (e^B -1)*100, gives pc change in Y from 1 unit change in X [http://www.cazaar.com/ta/econ113/interpreting-beta]
**Except with indicator variables, when the positive change is (e^B -1)*100 but the negative change is (e^(-B) -1)*100 [https://davegiles.blogspot.com/2011/03/dummies-for-dummies.html].
*Log-log: ((1.01^B)-1)*100 [https://www.cscu.cornell.edu/news/statnews/stnews83.pdf]
 
Note that some guides appear to contain errors [https://stats.idre.ucla.edu/other/mult-pkg/faq/general/faqhow-do-i-interpret-a-regression-model-when-some-variables-are-log-transformed/]!

Navigation menu