Path: news.mathworks.com!not-for-mail
From: Peter Perkins <Peter.Perkins@MathRemoveThisWorks.com>
Newsgroups: comp.soft-sys.matlab
Subject: Re: to convert non-normal distribution into normal
Date: Mon, 06 Jul 2009 11:46:02 -0400
Organization: The MathWorks, Inc.
Lines: 13
Message-ID: <h2t67q$miu$1@fred.mathworks.com>
References: <h2mgc1$3v1$1@fred.mathworks.com>
NNTP-Posting-Host: perkinsp.dhcp.mathworks.com
Mime-Version: 1.0
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: 7bit
X-Trace: fred.mathworks.com 1246895162 23134 172.31.57.88 (6 Jul 2009 15:46:02 GMT)
X-Complaints-To: news@mathworks.com
NNTP-Posting-Date: Mon, 6 Jul 2009 15:46:02 +0000 (UTC)
User-Agent: Thunderbird 2.0.0.22 (Windows/20090605)
In-Reply-To: <h2mgc1$3v1$1@fred.mathworks.com>
Xref: news.mathworks.com comp.soft-sys.matlab:553135


Adhithya Plato wrote:

> I will pleased to know, whether there is any command in MATLAB to convert, non-normal data into normal data.....but without changing the mean and standard deviation more than 1 to 2% ?

Stop worrying about the mean and the variance.  You can adjust those however you want.

You've posted at least twice about this.  Here, you seem to have asked for a transformation; earlier you described computing the mean and variance and generating samples from a normal distribution.  If your data are not normal, then simply fitting a normal to them isn't going to help with anything.  It is often possible to transform data to normality.  The simplest, most common transformation is the log, and Box-Cox transformations are also common.  A Johnson transformation (see JOHNSRND in the Statistics Toolbox) might also be helpful.

But what transformation you would need depends entirely on what you data look like.  You will need to experiment on your own.

But step back a minute.  You need to decide what you're trying to accomplish, and what you are going to do with the transformation.

Hope this helps.