[R] tests of lognormal distribution

From: Hyunchul Kim <sundol313_at_naver.com>
Date: Fri, 15 Feb 2008 15:33:48 +0900


Hi all,

I have a data of lognormal distribution (sample size > 1,000,000).

What I want to do is
1) to test if my dataset is a lognormal distribution or not (Histogram shows a nice normal distribution in log scale but I want to check) 2) two subsets from this dataset have same mean or not (like "t test" of normal distribution)

What I tried are
1) ad.test{truncgof} of R, which is a "Supremum Class Anderson-Darling test" but I got error message.

What I had problems are
1) Memory error happen with error message of "cannot allocate vector of size 1.0GB" and I cannot increase the computer spec. 2) I don't know what is "t test" of lognormal.

What I am thinking is
1) Is it OK to convert the data in log scale and do perform t test to say these dataset has same mean?

How can I solve my problem?
Thank you for any comments.

Hyunchul

        [[alternative HTML version deleted]]



R-help_at_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. Received on Fri 15 Feb 2008 - 06:36:28 GMT

Archive maintained by Robert King, hosted by the discipline of statistics at the University of Newcastle, Australia.
Archive generated by hypermail 2.2.0, at Fri 15 Feb 2008 - 07:30:14 GMT.

Mailing list information is available at https://stat.ethz.ch/mailman/listinfo/r-help. Please read the posting guide before posting to the list.

list of date sections of archive