Hello, I would like to know the diffrence between 2 commands : cor_auto (from qgraph package) and lavCor (from lavaan package) to compute a polychoric correlation matrix in order to do a network analysis. I have the responses to the SF-36 questionnaire (36 items with ordered responses) and I would like to have a polychoric correlation matrix by following the Sacha Epskamp's method. But I have a warning if using cor_auto : "Correlation matrix is not positive definite. Finding nearest positive definite matrix". I can ignore this warning or not ? Because I try to do a polychoric matrix in STATA software and I do not have a warning message. And I don't understand the difference between lavCor command, because When I use lavCor, I do not have the warning message either. //cor.auto1 <- cor_auto(Data_1it) //cor.lav1<-lavCor(Data_1it, ordered=names(Data_1it)) Thank you in advance for your answer. Line
[[alternative HTML version deleted]] ______________________________________________ R-help@r-project.org mailing list -- To UNSUBSCRIBE and more, see 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.