Re: [amelia] error: chol(): failed to converge & reproducibility

Steve, It sounds like you are writing: Whereas, for your purposes, you want: There is no "seed" argument in the amelia function. For example, compare the following code examples to see if this is what you wanted: ## library(Amelia) data<-data.frame( runif(50), rnorm(50) ) data[2:5,1]<-NA out11<-amelia(data,m=5,seed=2345, iCanMakeUpAnyArgumentNameILike=TRUE) out12<-amelia(data,m=5,seed=2345, iCanMakeUpAnyArgumentNameILike=TRUE) print(identical(out11$imputations[[1]], out12$imputations[[1]])) set.seed(2345) out21<-amelia(data,m=5) set.seed(2345) out22<-amelia(data,m=5) print(identical(out21$imputations[[1]], out22$imputations[[1]])) best, James -- James Honaker, Senior Research Scientist //// Institute for Quantitative Social Science, Harvard University ________________________________ From: amelia-bounces(a)lists.gking.harvard.edu [amelia-bounces(a)lists.gking.harvard.edu] on behalf of Matt Blackwell [m.blackwell(a)rochester.edu] Sent: Tuesday, June 11, 2013 11:04 AM To: Stephen Martin Cc: amelia(a)lists.gking.harvard.edu Subject: Re: [amelia] error: chol(): failed to converge & reproducibility Hi Steve, 1) The number of iterations will vary from imputation to imputation within runs of amelia(). This is important because it means that you will get different imputed values across the "m" imputations. The seed should make the entire run of amelia() the same as the next run of amelia(). Thus, "Imputation 2" in one run should have the same imputed values as "Imputation 2" in the next run with the same seed. 2) This is probably due to highly collinear variables in the dataset or a linear dependence in the data. You might check for a saturated set of dummy variables or linear transformations in the data and set those to "idvars". Hope that helps! Cheers, matt. ~~~~~~~~~~~ Matthew Blackwell Assistant Professor of Political Science University of Rochester url: http://www.mattblackwell.org On Tue, Jun 11, 2013 at 10:59 AM, Stephen Martin <stephen.martin@york.ac.uk<mailto:stephen.martin@york.ac.uk>> wrote: Hello, I am using Amelia II and have a couple of queries. (1) I would like to ensure the reproducibility of the imputations so I include 'seed=2345' in the Amelia line command. However, I notice that, if I repeat the run with exactly the same line command, the chain lengths differ at each imputation. Will I be getting the same imputations? (2) At the end of an imputation (with no parameters significantly changed since the last iteration), I sometimes get the message: error: chol(): failed to converge together with a little window headed 'Microsoft Visual C++ Runtime library' and the message ' This application has requested the Runtime to terminate it in an unusual way. Please contact the application's support team for more information.' How should I interpret and respond to this? Thanks. Steve -- Amelia mailing list served by HUIT [Un]Subscribe/View Archive: http://lists.gking.harvard.edu/?info=amelia More info about Amelia: http://gking.harvard.edu/amelia Amelia mailing list Amelia@lists.gking.harvard.edu<mailto:Amelia@lists.gking.harvard.edu> To unsubscribe from this list or get other information: https://lists.gking.harvard.edu/mailman/listinfo/amelia