Notice: On March 31, it was **announced** that Statalist is moving from an email list to a **forum**. The old list will shut down on April 23, and its replacement, **statalist.org** is already up and running.

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

From |
Wameq Raza <wameq.r@gmail.com> |

To |
statalist@hsphsun2.harvard.edu |

Subject |
st: Generating tables with predicted probabilities (asclogit) |

Date |
Fri, 28 Sep 2012 13:47:49 +0200 |

Hi everyone, I've been using STATA 12 to run a conditional logit model with alternative specific and case specific variables using the asclogit command. I've been able to run everything fine, but I'm trying to generate graphs of probabilities for the alternative specific variables per the categories in my dependent var. Just to give you a background, I've been modeling the determinants of health seeking behaviour and have 5 categories of providers (Public, Private, RMPs, Pharmacists and Others) as my dependent var and have avg cost, distance and waiting time as alt specific vars along with a battery of case specific vars such as demographics, income and so forth. At this point, what I want to do is graph the effect of say a change in cost on probabilities of choosing each of the providers. In a MNL model, one would predict probability of each of the equations and then use: predict p1, equation (#1) graph twoway (qfit p1 income) For the asclogit model, I've predicted the probability using predict p1, pr But I'm guessing this is predicting the probabilities for the entire model as opposed to based on provider choices (ie, specific equations)? Can anyone suggest anything I might be able to do? Best Wameq -- W A M E Q R A Z A * * For searches and help try: * http://www.stata.com/help.cgi?search * http://www.stata.com/support/faqs/resources/statalist-faq/ * http://www.ats.ucla.edu/stat/stata/

- Prev by Date:
**RE: st: How to list and make copies of .do files on a hard drive** - Next by Date:
**Re: st: Appropriate modelling - testing which set of exposures are more important** - Previous by thread:
**st: Cox model with unobserved heterogeneity** - Next by thread:
**st: basic panel regression RE FE FEVD** - Index(es):