Dear all, How do I make R show the R-squared (deviance explained by the model) in a logistic regression? Below is how I write my syntax. Basically I want to investigate density-dependence in parasitism of larvae. Note that in the end I perform a F-test because the dispersion factor (residual deviance / residual df) is significantly higher than 1. But how do I make R show the "R-squared"? Best wishes Johan> y<-cbind(para,unpara) > model<-glm(y~log(larvae),binomial) > summary(model)Call: glm(formula = y ~ log(larvae), family = binomial) Deviance Residuals: Min 1Q Median 3Q Max -2.0633 -1.6218 -0.1871 0.7907 2.7670 Coefficients: Estimate Std. Error z value Pr(>|z|) (Intercept) 1.0025 0.7049 1.422 0.15499 log(larvae) -1.0640 0.3870 -2.749 0.00597 ** (Dispersion parameter for binomial family taken to be 1) Null deviance: 35.981 on 12 degrees of freedom Residual deviance: 27.298 on 11 degrees of freedom AIC: 40.949 Number of Fisher Scoring iterations: 4> anova(model,test="F")Analysis of Deviance Table Model: binomial, link: logit Response: y Terms added sequentially (first to last) Df Deviance Resid. Df Resid. Dev F Pr(>F) NULL 12 35.981 log(larvae) 1 8.683 11 27.298 8.6828 0.003212 **
On Ti, 2005-03-29, 10:56, Johan Stenberg skrev:> How do I make R show the R-squared (deviance explained by the model) in > a logistic regression?Several definitions of R^2 exists in the GLM case. See e.g. Menard, S. (2000) Coefficients of determination for multiple logistic regression analysis. American Statistician, 54, 17-24. Mittlbock, M. and Schemper, M. (2002) Explained variation for logistic regression - small sample adjustments, confidence intervals and predictive precision. Biometrical Journal, 44, 263-272. Liao, J.G. and McGee, D. (2003) Adjusted coefficients of determination for logistic regression. American Statistician, 57, 161-165. IIRC, the last paper contains R code. HTH, Henric
Johan Stenberg wrote:> Dear all, > > How do I make R show the R-squared (deviance explained by the model) in > a logistic regression? > > Below is how I write my syntax. Basically I want to investigate > density-dependence in parasitism of larvae. Note that in the end I > perform a F-test because the dispersion factor (residual deviance / > residual df) is significantly higher than 1. But how do I make R show > the "R-squared"? > > Best wishes > JohanThe proportion of deviance explained has been shown to not be such a good measure. You can use the lrm function in the Design package to get various measures including ROC area (C index), Somers' Dxy and Kendall tau rank correlation, Nagelkerke generalization of R-squared for maximum likelihood-based models (related to Maddala and others). Frank Harrell> > >>y<-cbind(para,unpara) >>model<-glm(y~log(larvae),binomial) >>summary(model) > > > Call: > glm(formula = y ~ log(larvae), family = binomial) > > Deviance Residuals: > Min 1Q Median 3Q Max > -2.0633 -1.6218 -0.1871 0.7907 2.7670 > > Coefficients: > Estimate Std. Error z value Pr(>|z|) > (Intercept) 1.0025 0.7049 1.422 0.15499 > log(larvae) -1.0640 0.3870 -2.749 0.00597 ** > > (Dispersion parameter for binomial family taken to be 1) > > Null deviance: 35.981 on 12 degrees of freedom > Residual deviance: 27.298 on 11 degrees of freedom > AIC: 40.949 > > Number of Fisher Scoring iterations: 4 > > >>anova(model,test="F") > > Analysis of Deviance Table > > Model: binomial, link: logit > > Response: y > > Terms added sequentially (first to last) > > > Df Deviance Resid. Df Resid. Dev F Pr(>F) > NULL 12 35.981 > log(larvae) 1 8.683 11 27.298 8.6828 0.003212 ** > > ______________________________________________ > R-help at stat.math.ethz.ch mailing list > https://stat.ethz.ch/mailman/listinfo/r-help > PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html >-- Frank E Harrell Jr Professor and Chair School of Medicine Department of Biostatistics Vanderbilt University