Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: st: Discrete time hazard model-interval censored
From
Nyasha Tirivayi <[email protected]>
To
[email protected]
Subject
Re: st: Discrete time hazard model-interval censored
Date
Fri, 27 Aug 2010 22:06:18 +0200
Dear Steve
Thanks for the link. I will look at it.
Regards
Nyasha Tirivayi
Maastricht university
On Fri, Aug 27, 2010 at 9:57 PM, Steve Samuels <[email protected]> wrote:
> It appears that you've asked different versions of this question
> before. I'm no better equipped to answer it then the original
> responders. Take a look at this thread on estimating "average
> treatment effects" and Austin Nichol's references therein:
> http://www.stata.com/statalist/archive/2010-08/msg01268.html. I don't
> know what do if employed status at the start is endogenous.
>
> Best wishes,
>
> Steve
>
> On Fri, Aug 27, 2010 at 1:49 PM, Nyasha Tirivayi <[email protected]> wrote:
>> Dear Steve
>>
>> Thanks for your response. If i use a logistic regression for
>> transition to employment, do I follow the example below:
>>
>> logistic employment beneficiary age gender if LagEmployment==0
>>
>> where LagEmployment==0 is previous state of unemployment and
>> beneficiary indicates participating in a social program. Do I need to
>> xtset the data? And how do I correct for endogeneity of beneficiary
>> and LagEmployment?
>>
>> May you kindly advise
>>
>> Regards
>>
>> Nyasha Tirivayi
>> PhD Researcher
>> Maastricht University
>> Netherlands
>>
>> On Fri, Aug 27, 2010 at 7:37 PM, Steve Samuels <[email protected]> wrote:
>>> --
>>>
>>> With only one interval, a discrete hazard model is _not_ suitable.
>>> Use a -logistic- (or -cloglog-) model instead.
>>>
>>> Steve
>>>
>>> Steven J. Samuels
>>> [email protected]
>>> 18 Cantine's Island
>>> Saugerties NY 12477
>>> USA
>>> Voice: 845-246-0774
>>> Fax: 206-202-4783
>>>
>>> On Fri, Aug 27, 2010 at 1:26 PM, Nyasha Tirivayi <[email protected]> wrote:
>>>> Dear All
>>>>
>>>> I have a dataset on labour force participation in an African country.
>>>> There are two waves of interviews, January 2009 and August 2009. I am
>>>> interested in looking at the effects of a social program on the
>>>> transitions into employment. The survey asked questions like, "are you
>>>> empoyed at the moment" . I am especially interested in those who at
>>>> first interview had were not employed. However ,we do not know exactly
>>>> the date when they do get employed, we only have a response at second
>>>> interview whether they are employed or not.
>>>>
>>>> Could anyone advise me if a discrete time hazard model is suitable?
>>>> What stata methods do I use for the interval censoring? And how do I
>>>> correct for the selection bias of the social program?
>>>>
>>>> May you kindly advise
>>>>
>>>> Regards
>>>>
>>>> Nyasha Tirivayi
>>>> PhD researcher
>>>> Maastricht Graduate School of Governance
>>>> Maastricht University
>>>> Netherlands
>>>> *
>>>> * For searches and help try:
>>>> * http://www.stata.com/help.cgi?search
>>>> * http://www.stata.com/support/statalist/faq
>>>> * http://www.ats.ucla.edu/stat/stata/
>>>>
>>>
>>> *
>>> * For searches and help try:
>>> * http://www.stata.com/help.cgi?search
>>> * http://www.stata.com/support/statalist/faq
>>> * http://www.ats.ucla.edu/stat/stata/
>>>
>>
>> *
>> * For searches and help try:
>> * http://www.stata.com/help.cgi?search
>> * http://www.stata.com/support/statalist/faq
>> * http://www.ats.ucla.edu/stat/stata/
>>
>
> *
> * For searches and help try:
> * http://www.stata.com/help.cgi?search
> * http://www.stata.com/support/statalist/faq
> * http://www.ats.ucla.edu/stat/stata/
>
*
* For searches and help try:
* http://www.stata.com/help.cgi?search
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/