I am trying to compute the standard errors of an estimate obtained using the Lubotsky-Wittenberg (2006) methodology for proxy combination. Here it is the list of commands:
gen a=.
gen b=.
gen c=.
gen d=.
gen e=.
gen f=.
reg y x_1 x_2 x_3
replace a = _b[x_1]
replace b = _b[x_2]
replace c = _b[x_3]
ivreg x_1 ( x_1 = y)
replace d = _b[x_1]
ivreg x_2 ( x_1 = y)
replace e = _b[x_1]
ivreg x_3 ( x_1 = y)
replace f = _b[x_1]
qui gen pbhat =a*d + b*e + c*f
sum pbhat
The estimate is "pbhat" that is a linear combination of multivariate regression coefficients and IV regression coefficients. Can you help me in obtaining the standard errors with this procedure?
Thank you in advance.
Related Posts with standard errors when using Lubotsky-Wittenberg methodology for proxy combination
log-log regressionHi, i am trying to find out if it is possible to use the natural log of an interaction variable in a…
stcoxcal after multiple imputationHi, I have a dataset of 7,000 patients but multiple observations were missing (around 20-30% of the…
Constant term in pvarDear all, I'm estimating a panel vector autoregression model in Stata using the package pvar, and I…
Problem for append datasetsHello, I'm new with STATA, and i'm trying to perform an append of different datasets and there is a…
Is it possible to produce tables with rounded numbers in the Results window?I would like for numbers displayed in Stata's Results window to be rounded. To use an example, I'm e…
Subscribe to:
Post Comments (Atom)
0 Response to standard errors when using Lubotsky-Wittenberg methodology for proxy combination
Post a Comment