How to run a time series model on panel data?

[Mihailo Savic]

Finally,

Everything works perfectly! Thank you so much! This is the final product:

Code: Select all

matrix ((@rows(gvkey))/38,7) eq1_m matrix ((@rows(gvkey))/38,9) eq2_m matrix ((@rows(gvkey))/38,11) eq3_m matrix ((@rows(gvkey))/38,9) eq4_m matrix ((@rows(gvkey))/38,2) ys_adf scalar rows=@rows(gvkey) scalar i=1 scalar j=1 while i<rows !gvk=gvkey(i) smpl if gvkey=!gvk series y{!gvk}=log(adjusted_oibdp) series x{!gvk}=log(capital_expenditures) series s{!gvk}=sale series u{!gvk}=cur equation eq{!gvk}.ls y{!gvk} c y{!gvk}(-1) eq1_m(j,1) = !gvk eq1_m(j,2) = eq{!gvk}.@rbar2 eq1_m(j,3) = eq{!gvk}.@se eq1_m(j,4) = eq{!gvk}.@coefs(1) eq1_m(j,5) = eq{!gvk}.@tstats(1) eq1_m(j,6) = eq{!gvk}.@coefs(2) eq1_m(j,7) = eq{!gvk}.@tstats(2) equation eq{!gvk}x.ls y{!gvk} c y{!gvk}(-1) x{!gvk} eq2_m(j,1) = !gvk eq2_m(j,2) = eq{!gvk}x.@rbar2 eq2_m(j,3) = eq{!gvk}x.@se eq2_m(j,4) = eq{!gvk}x.@coefs(1) eq2_m(j,5) = eq{!gvk}x.@tstats(1) eq2_m(j,6) = eq{!gvk}x.@coefs(2) eq2_m(j,7) = eq{!gvk}x.@tstats(2) eq2_m(j,8) = eq{!gvk}x.@coefs(3) eq2_m(j,9) = eq{!gvk}x.@tstats(3) equation eq{!gvk}tr.ls y{!gvk} c y{!gvk}(-1) x{!gvk} @trend eq3_m(j,1) = !gvk eq3_m(j,2) = eq{!gvk}tr.@rbar2 eq3_m(j,3) = eq{!gvk}tr.@se eq3_m(j,4) = eq{!gvk}tr.@coefs(1) eq3_m(j,5) = eq{!gvk}tr.@tstats(1) eq3_m(j,6) = eq{!gvk}tr.@coefs(2) eq3_m(j,7) = eq{!gvk}tr.@tstats(2) eq3_m(j,8) = eq{!gvk}tr.@coefs(3) eq3_m(j,9) = eq{!gvk}tr.@tstats(3) eq3_m(j,10) = eq{!gvk}tr.@coefs(4) eq3_m(j,11) = eq{!gvk}tr.@tstats(4) equation eq{!gvk}cur.ls y{!gvk} c s{!gvk} u{!gvk} eq4_m(j,1) = !gvk eq4_m(j,2) = eq{!gvk}cur.@rbar2 eq4_m(j,3) = eq{!gvk}cur.@se eq4_m(j,4) = eq{!gvk}cur.@coefs(1) eq4_m(j,5) = eq{!gvk}cur.@tstats(1) eq4_m(j,6) = eq{!gvk}cur.@coefs(2) eq4_m(j,7) = eq{!gvk}cur.@tstats(2) eq4_m(j,8) = eq{!gvk}cur.@coefs(3) eq4_m(j,9) = eq{!gvk}cur.@tstats(3) eq{!gvk}cur.makeresid rescur freeze(adftable) y{!gvk}.uroot(adf) ys_adf(j,1)= !gvk ys_adf(j,2)= @val(adftable(7,4)) d adftable while i<rows and !gvk=gvkey(i) i=i+1 wend j=j+1 wend

Apart from that I still need to manually save the data into excel from the 5 matrixes and I still need to look at an individual eq to view the residuals (that were fixed by changing "scalar gvk" to "!gvk").

Thank you again. I hope I wont need further help. (But I probably will.)

Mihailo.