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

From |
Sophia Rüster <sophia.ruester@tu-dresden.de> |

To |
<statalist@hsphsun2.harvard.edu> |

Subject |
st: Marginal Effect Probit - percent points? |

Date |
Mon, 31 Aug 2009 16:36:53 +0200 |

Dear all, I have run a Probit model and calculated the marginal effects afterwards. Assuming a continuous RHS variable, the marginal effect is interpreted the following way: "An infinitesimal change in the RHS variable changes the probability that the endogenous variable takes the value of one by X%". Are these X% (1) percent points or (2) percent of the actual value? Example: Pr(Y=1) = 0.5 Marginal Effect of x on y = 0.1 (1) Does it mean Pr(Y=1) changes to 0.6 (0.5+0.1) (2) or does it mean Pr(Y=1) changes to 0.55 (0.5*1.1) Thanks a lot for a brief response! Sophie * * 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/

**Follow-Ups**:**st: RE: Marginal Effect Probit - percent points?***From:*"Villa Lora, Juan Miguel" <JUANMIGUELV@Contractual.iadb.org>

**Re: st: Marginal Effect Probit - percent points?***From:*Maarten buis <maartenbuis@yahoo.co.uk>

- Prev by Date:
**st: sts graph on a Mac** - Next by Date:
**Re: st: Snow Leopard breaks Stata file util commands** - Previous by thread:
**st: sts graph on a Mac** - Next by thread:
**Re: st: Marginal Effect Probit - percent points?** - Index(es):

© Copyright 1996–2015 StataCorp LP | Terms of use | Privacy | Contact us | What's new | Site index |