Re: [R] strange error message when using rnorm and rbinom

2016-06-14 Thread Jim Lemon
Hi JI, The most likely problems are negative numbers for sd or "k" being larger than the number of mu.m2 or disp.m2 values. Jim On Wed, Jun 15, 2016 at 4:06 AM, JI Cho wrote: > Dear R users, > > I have been using rnorm, rbinom and have been getting the following warning > message when I do not h

Re: [R] strange error message when using rnorm and rbinom

2016-06-14 Thread Bert Gunter
Do not believe everything you read on the internet! "In one of the resources I found in over the internet explains that the mean and sd should be declared as integers." That is compete crap. The mean of a norma/Gaussian can be any real; the sd can be any positive real. Moreover, one does not nee

Re: [R] strange error message when using rnorm and rbinom

2016-06-14 Thread Tom Wright
As you probably already guessed we are going to need to see the contents of nref, mu.m2 and disp.m2 to help. dput(nref) dput(mu.m2) dput(disp.m2) k might help too. On Tue, Jun 14, 2016 at 2:06 PM, JI Cho wrote: > Dear R users, > > I have been using rnorm, rbinom and have been getting the followi

[R] strange error message when using rnorm and rbinom

2016-06-14 Thread JI Cho
Dear R users, I have been using rnorm, rbinom and have been getting the following warning message when I do not have any NAs in my generated values. Warning messages:1: In rnorm(nref, mean = mu.m2[[k]], sd = sqrt(disp.m2[[k]])) : NAs produced In one of the resources I found in over the intern