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]

st: Re: st: Re: st: Re: st: Re: st: Re: st: RE: Truncated sample or Heckman selection‏

From   Maarten Buis <>
Subject   st: Re: st: Re: st: Re: st: Re: st: Re: st: RE: Truncated sample or Heckman selection‏
Date   Fri, 5 Oct 2012 17:31:01 +0200

On Fri, Oct 5, 2012 at 4:29 PM, Ebru Ozturk wrote:
> My question was whether I should use Heckman sample selection as I restricted my sample to innovating firms only.

Certainly not, to use Heckman selection you must include the
observations that are not selected. Even Heckman cannot do magic;
information on how the selected firms are different from not selected
firms has to come from somewhere... Since you don't want to use the
non-innovating firms you cannot use -heckman-.

> After discussing with you, I also confused as I said before many papers used tobit regression and did Heckman correction and never seen fractional logit model in these papers such as Grimpe & Kaiser, (2010); Cassiman & Veugelers, (2006).
> These papers are saying that, the data is left censored as it includes many zeros.
> Maybe I am not very good at statistics but when I read papers they use tobit regression. Whey then?

It is not uncommon that in a (sub-(sub-))discipline an error is
repeated many times by many people. So the fact that a method has been
used before does not proof that the method is correct. Instead you
need to understand the argument behind the methods and base your
decision on that.

> Also when I search for fractional logit model it says it is bounded between 0 and 1 but my dependent variable includes for instance, 0, 11, 13, 15, 25 values and so on.

Seems to me you have percentages instead of proportions. If that is
the case you can just divide your dependent variable by 100 to turn
percentages into proportions.

-- Maarten

Maarten L. Buis
Reichpietschufer 50
10785 Berlin
*   For searches and help try:

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