s-news
[Top] [All Lists]

Re: Fiiting Distributions to data

To: Pradeep John <pjohn@bslindia.com>
Subject: Re: Fiiting Distributions to data
From: Frank E Harrell Jr <f.harrell@vanderbilt.edu>
Date: Thu, 29 Jun 2006 06:58:58 -0500
Cc: s-news@lists.biostat.wustl.edu
In-reply-to: <008c01c69b35$4ca43640$da22a8c0@binarypdc.local>
References: <20060628230655.99DA31007139@mailgate.biostat.wustl.edu> <008c01c69b35$4ca43640$da22a8c0@binarypdc.local>
User-agent: Mozilla Thunderbird 1.0.7 (X11/20051013)
Pradeep John wrote:
Dear Users,

Could  you help me on the following querry.

I have three columns of data points. I want to fit all the statistical
distributions to these data points & would like to know which distribution
best fits my data.

I want to know how to proceed with that in S-Plus as i am a new user of
S-Plus 7.0

Any help on the above querry will be appreciated.

Thanks & Regards
John
Statistician

That approach will result in a fitted cumulative distribution whose true precision equals that of the empirical cumulative distribution function. In that case it's much easier just to use the ECDF to describe the data.

Frank Harrell

<Prev in Thread] Current Thread [Next in Thread>