Hello all, I am running a simple path analysis with the function sem.mi (of semTools) after doing multiple imputation in my (missing) data. However, depending on the option to combine the chi-square, I get the following warning messages:
Warning messages: 1: In estimateVCOV(lavaanModel, samplestats = lavaanSampleStats, ... : lavaan WARNING: could not compute standard errors! 2: In pchisq(chisq, df) : NaNs produced 3: In estimateVCOV(lavaanModel, samplestats = lavaanSampleStats, ... : lavaan WARNING: could not compute standard errors! 4: In pchisq(chisq, df) : NaNs produced 5: In estimateVCOV(lavaanModel, samplestats = lavaanSampleStats, ... : lavaan WARNING: could not compute standard errors! and so forth. The options chi="mr" and "mplus" result in these warning messages, but options chi="lmrr" and "none" run fine (no warning messages). Even when I get these warning messages, all estimates (including se and chi) are printed out in the results (using summary, for example). Also, using the function sem (of lavaan package) directly (with one of the "replicated" datasets) runs fine. Here is the code I'm using: # start code # model syntax model1 <- ' # regressions r1 ~ p1+p2+p3+p4+p5+p6+p7+p8+p9+p10+p11 r2 ~ r1+p2+p4+p5+p6+p8+p9+p10+p11+p12+p13+p14 ' # run sem (N=124); data are already imputed out1 <- sem.mi(model1,imputedData,m=20,chi="mr",fixed.x=T,std.ov=T) summary(out1) inspect(out1, "imputed") # the combined chi is presented # end code Can someone tell me why am I getting these warning messages? Thanks, Duarte [[alternative HTML version deleted]] ______________________________________________ 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.