Hello All,
I try to compare market reactions based on earnings as compared to contract announcements. I have two samples with the same number of observations.
1. Rit (annual)=a0+ b1REit (w), where
- Rit (annual) - calendar-year buy-and-hold returns of firm i in year t
- b1REi - returns of firm i in year t in earnings announcement window
2. Rit (annual)=a0+ b1RCit (w), where
- Rit (annual) - calendar-year buy-and-hold returns of firm i in year t
- b1RCi - returns of firm i in year t in contract announcement window
I then compare adjusted R2 from both regressions, but it is not enough. For the sake of reliability check, I want to apply Vuong to see that both models are “equally far away” from the data that is being modelled. Just need to know how to make the test in STATA.
Thank you in advance!!!
Kind regards,
Daria
Related Posts with The Vuong test for 2 non-nested simple linear regression models
Scatter plot with a categorical variableI have three categorical variables (capacity, times, prices). The capacity has three levels ( 4 cups…
Strong Correlation of Variables in Spearmans'rho MatrixHello Just before doing a panel regression, I wanted to look at the correlation matrix for the desc…
The krls command and panel data regressionsHello Forum members, I have come across the krls command for Kernel-Based Regularized Least Squares…
How to conduct multiple impulation when all variables have missing values?Dear all, I am working with my resident survey data (n=8356) including 59 items, most of which are …
SEM Builder: Wrong modelHi, I am trying to run a first-order four-factor model in Stata 15 SEM Builder: Array However, the …
Subscribe to:
Post Comments (Atom)
0 Response to The Vuong test for 2 non-nested simple linear regression models
Post a Comment