Ordered probit model

[EViews Glenn]

Here's an example that replicates the computation. Note that the Makepcomp default is to use the correlation (the docs are wrong on this one -- I'll get them corrected) so that we have to scale the Y before use. At the end of this, the series in A1, A2, A3, matches the matrix of scores in SCORES. I just grabbed the series LBLA, etc. for this example; you should substitute your own.

As noted in the manual

Code: Select all

Note that when computing scores using Equation (12.33), EViews will transform the Y to match the data used in the original computation. For example, the data will be scaled for analysis of correlation matrices, and partialing will remove means and any conditioning variables

group group01 lbla loan oil orders
group01.makepcomp(eigvec=evec) a1 a2 a3
group group02 a1 a2 a3

group group03 (lbla-@mean(lbla))/@stdevp(lbla) (loan-@mean(loan))/@stdevp(loan) (oil-@mean(oil))/@stdevp(oil) (orders-@mean(orders))/@stdevp(orders)
stom(group03, ff)

matrix scores = ff*evec

[rocky]

Dear All,
I am using E-views 6, have calculated the probit and logit model. Now, I am not sure how to calculate its Marginal effects. My dependent variable is 1,0 (Y) variable. Despite reading the above discussion, I am still not able to solve the problem. I have also used R to calculate Marginal Effects, but the problem is that the McFadden R-Square in R for the same data and equation is very high as compared to what I calculate in eViews (i.e: in eViews its 7% and R shows me this to be 86%).
My Model looks like this: Y c V1 v2 V3 V4 V5 V6 V7 V8 V9
If your answer includes equations starting from ''@'', please tell me where I have to write such type of equations as I am new to eViews.
Thanks
Rocky...