Dear all,
I am using Stata13 and specified a recursive model of this form:
BC = a0+aiX (where X is a vector of financial variables)+e.......(1)
Inc = b0+b1BC+b2Z (where Z is a vector of control variables)+u......(2)
Since this is a time series data, I have the following questions:
1) do I have to undertake stationarity tests for all the variables?
2) will I have to use the residual or predicted value of BC in equation 2?
3) can this recursive model mitigate endogeneity?
I will appreciate any form of progressive contribution that will point me in te right direction and maybe a Stata do-file (that can be modified)....that will really help!
Anyone please?
Thanks in advance!