Bookmark and Share

Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: st: Data structure

From   Nick Cox <>
Subject   Re: st: Data structure
Date   Sun, 17 Jun 2012 02:07:10 +0100

It's best considered as continuous, in my view.

It's likely that Poisson regression will make good sense, however. See

On Sun, Jun 17, 2012 at 1:56 AM, ming li <> wrote:

> I'm working on a paper about people's donation behavior after an
> earthquake. Most of the dependent variable (total amount donated)
> clusters at values like 50, 100, 200, 300, 400, 500, etc. I think this
> is largely caused by currency values. Therefore it seems not
> appropriate to treat it as a continuous variable. Do you have any
> suggestions about what should I do when dealing with such kind of
> data?
*   For searches and help try:

© Copyright 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index