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

st: endogenous switching regression

From   "fmodena" <>
To   <>
Subject   st: endogenous switching regression
Date   Mon, 8 Oct 2007 16:37:24 +0200

Dear all,

I am looking at the effects of a crop loss and the labour supply response to
the crop loss on household income in rural Indonesia. The household income
is regressed on the dummy crop loss (CL) interacted  with the value of farm
assets and on the dummy labour supply as a response to the crop loss (LS)
interacted with the value of farm assets. Note that LS (labour supply
response) is observed only for those who had the crop loss (CL=1), but I
replaced LS=0 if CL=0. 
The estimated crop loss effect and labour supply effect are then included in
the expenditure equation, to test weather households reduce consumption in
the face of a crop loss, and weather the labour supply response to the shock
influences the decisions to cut expenditure. 

The labour supply response to a crop loss may be endogenous in the income
equation (least squares estimation may lead to biased estimates of the
parameter). I cannot use a treatment effect model (treatreg) to control for
the endogeneity of the labour supply because the dummy LS is interacted with
the farm assets (which is a continuous variable). Hence I followed the
procedure proposed by Cameron and Worswick (1999). The authors employ a
switching regression model with endogenous switching. This method involves a
two stage procedure:
1)	estimate a probit equation with the dummy LS as dependent variable
(using the sub-sample for which CL=1). Then calculate the two selection
terms (inverse Mills's ratio) for labour supply respondent (LS=1) and for
non labour supply respondents (LS=0)
2)	the two selection terms are included into the income equation (using
the entire sample)

I am not completely satisfied by that procedure (the two selection terms
seem not able to capture the endogeneity of LS in the income equation). I am
also wondering weather the movestay command could help me (instead of using
the two stage approach). 

Any assistance will be appreciated 

*   For searches and help try:

© Copyright 1996–2017 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index