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 |
Steven Samuels <sjhsamuels@earthlink.net> |

To |
statalist@hsphsun2.harvard.edu |

Subject |
Re: AW: st: AW: Fitted probabilities using prvalue for logit model |

Date |
Mon, 19 Jul 2010 10:20:50 -0400 |

Steve On Jul 19, 2010, at 8:53 AM, Marc Michelsen wrote:

will try to figure out which one is well suited for my kind of analysis. -----Ursprüngliche Nachricht----- Von: owner-statalist@hsphsun2.harvard.edu

Gesendet: Freitag, 16. Juli 2010 17:02 An: statalist@hsphsun2.harvard.edu Betreff: Re: st: AW: Fitted probabilities using prvalue for logit model I don't know what you mean by "determine the relative importance of my additional dummy variables relative to the benchmark model with its explanatory variables?" But in this case -prvalue- is obviously not working for you. How are you measuring "importance"? If you mean "significance", have you tested the joint significance of the two variables with -test-? (Adding variables will always increase the log-likelihood, so "improvement" is not a guide). If the criterion of importance is "predictive accuracy", then compare ROC curves for the two models with -roccomp-. Unfortunately, the ROCs for both models will be systematically optimistic, but the differences could still be revealing. For better accuracy, some kind of cross-validation approach is needed. For cross-validation approaches, see: http://www.stata.com/statalist/archive/2008-02/msg00686.html An unreferenced Stata program for cross-validation is contained in: http://www.mail-archive.com/r-help@r-project.org/msg82508.html There is also a literature on "proportional reduction in error" approaches, including partial r-squares. See: Agrestic, Analysis of Categorical Data, 2nd Ed (2002) Wiley, Chapter 6. Measures of r-square based on the log-likelihood are difficult to interpret (p. 227). A Google search will turn up many references. (By the way, -prvalue- is not an official Stata command. I presume it is user-written. Please, as the FAQ request, give references for all the non-Stata commands you use.) Steve On Fri, Jul 16, 2010 at 9:54 AM, Marc Michelsen <marcmichelsen@t-online.de> wrote:

Steve,of course there are four possible combinations -- however, in my set-upthere are only three valid combinations. 1/1 is not possible.Does your statement mean that -prvalue- is not an appropriatemeasure todetermine the relative importance of my additional dummy variables

relative

to the benchmark model with its explanatory variables? Marc -----Ursprüngliche Nachricht----- Von: owner-statalist@hsphsun2.harvard.edu[mailto:owner-statalist@hsphsun2.harvard.edu] Im Auftrag von SteveSamuelsGesendet: Freitag, 16. Juli 2010 15:18 An: statalist@hsphsun2.harvard.eduBetreff: Re: st: AW: Fitted probabilities using prvalue for logitmodelThere are four combinations of two dummy variables, not three, so your statements don't make sense. The coefficients of the variables, if you hold others constant, refer only to the relative associations among those four categories, not to any absolute levels. Those are determined by the values at which you fix the other covariates and by the constant term. .It is well-known that prediction at the means of covariates will not even reproduce the mean prediction, which in turn is the raw prevalence. It is quite possible that all four predictions could be lower than the crude prevalence rate. So, there's no reason to expect those predictions to match those of any "benchmark" model and a (single?) benchmark probability. Steve On Fri, Jul 16, 2010 at 6:02 AM, Marc Michelsen <marcmichelsen@t-online.de> wrote:Dear all,as I didn't get an answer to my problem below, I am trying torewrite thequestion more precisely/generally. The reference for the approachis thefollowing: DeAngelo, H., L. DeAngelo, and R. M. Stulz. "Seasonedequityofferings, market timing, and the corporate lifecycle." Journal ofFinancialEconomics 95 (2009), 275-295. I am referring to the table on page284.I am estimating the fitted probabilities of a logit model at fixedlevelsofthe explanatory variables using -prvalue-. I have a benchmark modelandtherefore also a benchmark probability of the event. Including my twodummyvariables in a second model specification (improves Peusdo-R2 andChi2)actually lowers the probability of the event. However, theprobabilityshould increase if the dummy variables are coded 0 (dummy 1)/1(dummy 2).The probabilities are lower in all three possible combinations ofthe twodummies. Although the coefficients of the logit model show thecorrectsignsand are statistically significant for one of the dummy variables. Does anybody has a view on this? Many thanks for considering this posting Marc -----Ursprüngliche Nachricht----- Von: owner-statalist@hsphsun2.harvard.edu [mailto:owner-statalist@hsphsun2.harvard.edu] Im Auftrag von MarcMichelsenGesendet: Donnerstag, 15. Juli 2010 11:12 An: statalist@hsphsun2.harvard.edu Betreff: st: Fitted probabilities using prvalue for logit model Dear Statalist users,I am running a logit model to estimate the effect and relativeimportanceofmarket timing and rating concerns on the decision to conduct aseasonedequity offering (panel data).Including my rating concern proxy variables in the regressionsimprovesthefit of the logit model (Pseudo-R2 and Chi2) compared to the standard

model

(including only market timing and control variables). One of the tworatingconcern proxies (positive rating momentum) is statisticallysignificant

at

5% with a marginal effect of -1.7%. The other one (negative ratingmomentum)shows a positive marginal effect but has no significant influence.In order to gauge the relative importance of market timing versusratingconcerns, I am trying to obtain predicted probabilities ofconducting aseasoned equity offerings (SEO) in a given year. Therefore, I amusing

the

"prvalue" command to calculate the probabilities at representativevaluesofthe explanatory variables (control variables at sample means, goodvs.poormarket timing opportunities). Neutral market timing opportunitiestranslatesinto a SEO probability of 5.2%, which is comparable to the study vonDeAngelo/DeAngelo/Stulz (2009) p. 284. But if I measure theprobabilitiesfor positive, negative and neutral rating momentum (the otherexplanatoryvariables are set equal to the former model specification), theprobabilities are always lower compared to the benchmark model(3.8% /5.0%/ 4.9%). While it is reasonable to assume that positive ratingmomentumlower the SEO probability, the results for the two other ratingvariablesare surprising.Obviously, this weakens my hypothesis that rating concerns are oneof thedrivers of seasoned equity offerings.Does anybody have an idea why the fitted probabilities are lower inallthree cases although the model fit is improved if I include the

respective

explanatory variables? Many thanks Marc * * 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/ * * 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/-- Steven Samuels sjsamuels@gmail.com 18 Cantine's Island Saugerties NY 12477 USA Voice: 845-246-0774 Fax: 206-202-4783 * * 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/ * * 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/

-- Steven Samuels sjsamuels@gmail.com 18 Cantine's Island Saugerties NY 12477 USA Voice: 845-246-0774 Fax: 206-202-4783 * * 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/ * * 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/ * * 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/

**References**:**st: Fitted probabilities using prvalue for logit model***From:*"Marc Michelsen" <marcmichelsen@t-online.de>

**st: AW: Fitted probabilities using prvalue for logit model***From:*"Marc Michelsen" <marcmichelsen@t-online.de>

**Re: st: AW: Fitted probabilities using prvalue for logit model***From:*Steve Samuels <sjsamuels@gmail.com>

**AW: st: AW: Fitted probabilities using prvalue for logit model***From:*"Marc Michelsen" <marcmichelsen@t-online.de>

**Re: st: AW: Fitted probabilities using prvalue for logit model***From:*Steve Samuels <sjsamuels@gmail.com>

**AW: st: AW: Fitted probabilities using prvalue for logit model***From:*"Marc Michelsen" <marcmichelsen@t-online.de>

- Prev by Date:
**Re: Re: st: st. Simultaneous Equations Model & GMM Estimation** - Next by Date:
**st: Append in combination with insheet (of csv files)** - Previous by thread:
**AW: st: AW: Fitted probabilities using prvalue for logit model** - Next by thread:
**st: Nested Logit Model** - Index(es):