Hi!
I have tried to find the "best" variables to include in my psmatch2.
I want to match public firm to private firm (bors_aks), to see if there is any difference in the investment level (gross_inv_scaled).
1 ) If p>|z| is greater than 0.05, then we cannot reject the H0? H0 = no difference between the firms.
= conclusion: all of our variable is equal between public and private firm. Correct? and is it good?
2) "Note: 5 failures and 0 sucesses completely determined" => Is that good for our model? since we want to match on variables that is equal between public and private firms?
3) T-STAT value under ATT = 5.10 => the difference in gross_inv_scaled is significant?
4) Under the PSTEST: Does the bias look good? should it be less than <5 %? or is it less than < | 5% | ?
5) Should I try to include more variables in the psmatch2?
- I have tried to include firm size, profit, but the bias increases and the p-value reduces..
Thank!
I have tried to find the "best" variables to include in my psmatch2.
I want to match public firm to private firm (bors_aks), to see if there is any difference in the investment level (gross_inv_scaled).
1 ) If p>|z| is greater than 0.05, then we cannot reject the H0? H0 = no difference between the firms.
= conclusion: all of our variable is equal between public and private firm. Correct? and is it good?
2) "Note: 5 failures and 0 sucesses completely determined" => Is that good for our model? since we want to match on variables that is equal between public and private firms?
3) T-STAT value under ATT = 5.10 => the difference in gross_inv_scaled is significant?
4) Under the PSTEST: Does the bias look good? should it be less than <5 %? or is it less than < | 5% | ?
5) Should I try to include more variables in the psmatch2?
- I have tried to include firm size, profit, but the bias increases and the p-value reduces..
Thank!
Attached Files
