> > Dear Madam, Sir, > Dear Madam, Sir, > I am a student in statistics at the Université Rennes 1 (France), I've used > the polr function in R to estimate an ordered probit model (consumer > preference) but the strange thing is that *I did not get the probabilities > which can tell me whether the variables are significant or not*. My model > has a dep var with 6 levels (0 to 5) and 19 explanatory variables (both > dummy and continues). Below is the code that I used and the result that I > got. Also I enclosed a part of the data used in the model for further help. > I am waiting for your help to fix this issue, many thanks. > *******Code**** > m <- sqlQuery(channel = 1, select * from [Feuil2$]) > m$Depvar <- as.factor(m$Depvar) > library(MASS, pos=4) > prm <- polr(Depvar ~ var1 + var2 + var3 + var4 + var5 +var6 + var7 + var8 + > > var9, method="probit", data=m, Hess=TRUE) > summary(prm) > > *****Output***** > > m <- sqlQuery(channel = 1, select * from [Feuil2$]) > > m$Depvar <- as.factor(m$Depvar) > > library(MASS, pos=4) > > prm <- polr(Depvar ~ var1 + var2 + var3 + var4 + var5 +var6 + var7 + var8 > + > + var9, method="probit", data=m, Hess=TRUE) > > summary(prm) > Call: > polr(formula = Depvar ~ var1 + var2 + var3 + var4 + var5 + var6 + > var7 + var8 + var9, data = m, Hess = TRUE, method = "probit") > Coefficients: > Value Std. Error t value *how to get P-values > ? > *var1 0.04454384 0.10390217 0.4287094 *? > *var2 0.22172411 0.10221326 2.1692305 *?* > var3 -0.19858516 0.09909994 -2.0038879 *?* > var4 0.48806601 0.10375119 4.7041967 *?* > var5 0.01845784 0.10018124 0.1842444 *?* > var6 -0.37819633 0.10170134 -3.7186955 *?* > var7 0.06047232 0.03234284 1.8697284 *?* > var8 0.39855835 0.19211257 2.0746084 *?* > var9 0.16119371 0.04692679 3.4350044 *?* > Intercepts: > Value Std. Error t value > 0|1 -0.2063 0.2931 -0.7040 > 1|2 0.3162 0.2903 1.0889 > 2|3 0.7557 0.2900 2.6059 > 3|4 1.2209 0.2913 4.1915 > 4|5 1.7293 0.2944 5.8736 > Residual Deviance: 1667.097 > AIC: 1695.097 > Thank you for your time > sincerely > Thierno, Université de Rennes 1 France >
______________________________________________ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
