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

Re: st: conditions to use the maximum likelihood function

From   Maarten buis <[email protected]>
To   [email protected]
Subject   Re: st: conditions to use the maximum likelihood function
Date   Mon, 14 Sep 2009 16:35:20 +0000 (GMT)

--- On Mon, 14/9/09, amatoallah ouchen wrote:
> I'm using a maximum likelihood function to estimate the
> parameters of a tobit model, and I'want to know if there
> is any special conditions, any trap, I'have to pay
> attention to, such as, the minimal observations number?...

In principle the rules are that the model is exactly true
and if you want to report tests you also need to have an 
infinite number of observations. In practice these 
conditions are never satisfied, so then the question 
becomes "how bad is still acceptable" and not surprisingly
the answer is "it depends", most importantly on the model.
Some readable discussions of the -tobit- model can be 
found in (Breen 1996) and (Long 1997).

Richard Breen (1996) "Regression models: Censored, sample 
selected, or truncated data" Thousand Oaks: Sage.

J. Scott Long (1997) "Regression models for categorical
and limited dependent variables" Thousand Oaks: Sage.

Hope this helps,

Maarten L. Buis
Institut fuer Soziologie
Universitaet Tuebingen
Wilhelmstrasse 36
72074 Tuebingen


*   For searches and help try:

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