Statalist


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

st: Interpret Ln transormed in logit regression


From   "Supnithadnaporn, Anupit" <gtg065t@mail.gatech.edu>
To   statalist <statalist@hsphsun2.harvard.edu>
Subject   st: Interpret Ln transormed in logit regression
Date   Mon, 29 Dec 2008 20:05:52 -0500 (EST)

Dear all,

Would anybody please confirm or correct my understanding about the interpretation of these coefficients? Any reference would be appreciated.

Y = Pr(vehicle fail the inspection); coded 1 if fail and 0 if pass
X = ln(income)
Z = other control variables

logit(Y) = a + b_1*X + b_i*Z

Are all of these correct?
b_1 = elasticity = increase income will decrease the odds of Y by b_1
e^b_1 = odds ratio (OR) = odds of Y increase by a factor of [e^b_1] for every unit of ln(income)
(OR-1)*100 = percentage change in odds of Y per unit change of ln(income)

Thank you
Anupit


*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/



© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index