Statalist The Stata Listserver

[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

Re: st: Trimming data and statistics

From   "Akihito Tokuhara" <[email protected]>
To   [email protected]
Subject   Re: st: Trimming data and statistics
Date   Fri, 14 Apr 2006 23:21:35 -0500

Dear Maarten,

I understand what you explained, and that is partly why I said I am hesitated to interpret the result.

And partly, at least, in one context, if the majority of replies comes back at zero value, these values
might be considered (assumed) as some "constant" effect of some factors which may likely exist,
for example, errors, unresponse due to filters, etc. Therefore, under this assumption, if the sample
size is increased, the number of those values also increases, so the median won't change whether
the sample size changes or not. Now cutting part of the "constant" effect may reduce the
overwhelming impact of such "constant" effect so that the "responsive" part may be seen.

But even so, I tend to think that this argument is only suitable in very few contexts. If I am wrong,
it is not a surprise either.


From: Maarten buis <[email protected]>
Reply-To: [email protected]
To: [email protected]
Subject: Re: st: Trimming data and statistics
Date: Fri, 14 Apr 2006 22:21:11 +0100 (BST)

I am not surprised that the middel observation (median) moves if you remove the bottom 5% and the
top 10% of a distribution. The problem is how to justify it.

--- Akihito Tokuhara <[email protected]> wrote:
> The fact is, when I test-trim the dataset, the median actually
> moves away from zero.

--- Akihito Tokuhara <[email protected]> previously wrote:
>> My way if to cut off completely all obs having values less than p5 and all
>> those having values bigger than p90.

Maarten L. Buis
Department of Social Research Methodology
Vrije Universiteit Amsterdam
Boelelaan 1081
1081 HV Amsterdam
The Netherlands

visiting adress:
Buitenveldertselaan 3 (Metropolitan), room Z214

+31 20 5986715

Switch an email account to Yahoo! Mail, you could win FIFA World Cup tickets.
* For searches and help try:
Don�t just search. Find. Check out the new MSN Search!

* For searches and help try:

© Copyright 1996–2024 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index