I want to use the GMM technique to estimate parameters of Fama Fench three factor model. I don't have stata license so how can i do with Eviews? I saw there is GMM between the methods available but i don't understand what to write in the field "instruments". Has anyone ever used Eviews to apply GMM estimation technique? Thank You in advance.
You might try R. Tutorial here: https://www.youtube.com/watch?v=dc1L3zjsgiY
R, of course, is free.
You might try R. Tutorial here: https://www.youtube.com/watch?v=dc1L3zjsgiY
R, of course, is free.
Thank You very much for the valuable answers. I found how to perform the gmm with Eviews even if i have doubts about the way J-statistic is calculated.
Hello
Eviews contains the estimator of generalized moments. But, first, only the "old" version of Arrelano and Bond is available; the "generalized system time" version of Blundel and Bond is not available on Eviews. Then it is possible to estimate the parameters of your model on Eviews but you will have to know the instruments. The instruments are the variables that will replace the instrumented variable (or that which is supposed to correlate with the disturbances); If you do not know the instruments that will replace this variable, your model will be incorrectly specified. To reduce the risks of mis-specification, authors choose the delays of the variable to be instrumented. Finally, know that the GMM method is a first difference estimator, that is, it regresses the variation of one variable on the variation of another variable: does the Fama model determine the average cost of capital or its variation.
Zoubir Faiçal Hi,
I agree with You, i'm applying the procedure used by Brighi,D'Addona,Dalla Bina "The Determinants of Risk Premia on the Italian Stock Market: Empirical Evidence on Common Factors in Asset Pricing Models" Economic Notes, 2013, that you can find attached at page 109.
Yes, absolutely eviews is suitable for GMM, and run some estimates of GMM models. You have to test the instrumental variables in GMM model and the model validity. If the instrumental variables are strong and the model is valid, then you can use the estimated GMM model. However, if the model has weak instrumental variables and is not valid, OLS is superior to GMM
can use Eviews for GMM . R is also suitable if you want the latest of statistical test but be very sure of the biases as OLS does give better response when robustness is of concerns.
I would recommend the latest gmm package by Chausse (2018) in R. But if Eviews did a good job, then you may just compare your results with those in R.
The Fama-French 3-factor model depends critically on issues such as correct functional form, especially linearity, and exogeneity.
GMM does not resolve the problem of correct specification.
Dear, important things you most choices it good, instruments variables for using GMM.
1) Use R
2) Allerano and Bond (1991) estimation is good.
3) Specification error will always be there in method of moments, kindly check
- Badges
- Science topic
- Similar topics
- Mathematics
- Statistics