Jan Verbesselt
2005-Feb-22 14:19 UTC
[R] ERROR NaNs produced; when comparing two logistic regression models with the ANOVA CHI test
Dear R-list, *When comparing two logistic regression models with the anova CHi test, I obtain the following error: (there are no NA's in the time series). How can this be solved such that I can compare two models on the same dataset were different explanatory variables are used? l.KBDI <- glm(zna.arson2 ~ zna.KBDI,family = binomial) l.NDWI <- glm(zna.arson2 ~ zna.NDWI,family = binomial) anova(l.KBDI,l.NDWI, test = "Chi") Analysis of Deviance Table Model 1: zna.arson2 ~ zna.KBDI Model 2: zna.arson2 ~ zna.NDWI Resid. Df Resid. Dev Df Deviance P(>|Chi|) 1 110 123.227 2 110 113.593 0 9.635 Warning message: NaNs produced in: pchisq(q, df, lower.tail, log.p) *Is this caused by the non-linearity of the data? *How can I obtain a p-value? Thanks, Jan _______________________________________________________________________ ir. Jan Verbesselt Research Associate Lab of Geomatics and Forest Engineering K.U. Leuven Vital Decosterstraat 102. B-3000 Leuven Belgium Tel:+32-16-329750 Fax: +32-16-329760 http://gloveg.kuleuven.ac.be/
Prof Brian Ripley
2005-Feb-22 14:58 UTC
[R] ERROR NaNs produced; when comparing two logistic regression models with the ANOVA CHI test
You need the models to be NESTED for this test to be valid. Look up `analysis of deviance' in your textbooks. You get the error because you have two unnested models with the same number of parameters, so the F numerator is divided by 0. (But this is merely a symptom of incorrect usage.) On Tue, 22 Feb 2005, Jan Verbesselt wrote:> Dear R-list, > > *When comparing two logistic regression models with the anova CHi test, I > obtain the following error: (there are no NA's in the time series). How can > this be solved such that I can compare two models on the same dataset were > different explanatory variables are used? > > > l.KBDI <- glm(zna.arson2 ~ zna.KBDI,family = binomial) > l.NDWI <- glm(zna.arson2 ~ zna.NDWI,family = binomial) > > anova(l.KBDI,l.NDWI, test = "Chi") > > > Analysis of Deviance Table > > Model 1: zna.arson2 ~ zna.KBDI > Model 2: zna.arson2 ~ zna.NDWI > Resid. Df Resid. Dev Df Deviance P(>|Chi|) > 1 110 123.227 > 2 110 113.593 0 9.635 > Warning message: > NaNs produced in: pchisq(q, df, lower.tail, log.p) > > > *Is this caused by the non-linearity of the data? > *How can I obtain a p-value?-- Brian D. Ripley, ripley at stats.ox.ac.uk Professor of Applied Statistics, http://www.stats.ox.ac.uk/~ripley/ University of Oxford, Tel: +44 1865 272861 (self) 1 South Parks Road, +44 1865 272866 (PA) Oxford OX1 3TG, UK Fax: +44 1865 272595
Dimitris Rizopoulos
2005-Feb-22 15:02 UTC
[R] ERROR NaNs produced; when comparing two logistic regression models with the ANOVACHI test
these models are not nested and thus LRT should not be used. You get no p-value because Df=0 in your case. As an alternative you could look at the deviances. I hope it helps. Best, Dimitris p.s., you referred to `non-linearity of the data' if you mean that the probability of positive responses is a nonlinear function of your explanatory variables then you could use splines to model it (see e.g., at function `lrm()' from the Design package). ---- Dimitris Rizopoulos Ph.D. Student Biostatistical Centre School of Public Health Catholic University of Leuven Address: Kapucijnenvoer 35, Leuven, Belgium Tel: +32/16/336899 Fax: +32/16/337015 Web: http://www.med.kuleuven.ac.be/biostat/ http://www.student.kuleuven.ac.be/~m0390867/dimitris.htm ----- Original Message ----- From: "Jan Verbesselt" <Jan.Verbesselt at agr.kuleuven.ac.be> To: <r-help at stat.math.ethz.ch> Sent: Tuesday, February 22, 2005 3:19 PM Subject: [R] ERROR NaNs produced;when comparing two logistic regression models with the ANOVACHI test> Dear R-list, > > *When comparing two logistic regression models with the anova CHi > test, I > obtain the following error: (there are no NA's in the time series). > How can > this be solved such that I can compare two models on the same > dataset were > different explanatory variables are used? > > > l.KBDI <- glm(zna.arson2 ~ zna.KBDI,family = binomial) > l.NDWI <- glm(zna.arson2 ~ zna.NDWI,family = binomial) > > anova(l.KBDI,l.NDWI, test = "Chi") > > > Analysis of Deviance Table > > Model 1: zna.arson2 ~ zna.KBDI > Model 2: zna.arson2 ~ zna.NDWI > Resid. Df Resid. Dev Df Deviance P(>|Chi|) > 1 110 123.227 > 2 110 113.593 0 9.635 > Warning message: > NaNs produced in: pchisq(q, df, lower.tail, log.p) > > > *Is this caused by the non-linearity of the data? > *How can I obtain a p-value? > > > Thanks, > Jan > > _______________________________________________________________________ > ir. Jan Verbesselt > Research Associate > Lab of Geomatics and Forest Engineering K.U. Leuven > Vital Decosterstraat 102. B-3000 Leuven Belgium > Tel:+32-16-329750 Fax: +32-16-329760 > http://gloveg.kuleuven.ac.be/ > > ______________________________________________ > 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 >
Apparently Analagous Threads
- negetative AIC values: How to compare models with negative AIC's
- labels on axis(4) and adaptation of legend to size of the plot
- Penalized likelihood-ratio chi-squared statistic: L.R. model for Goodness of fit?
- Change labels of x-axes in Plot of stl() function?
- Spatial Autocorrelation for point data