I need to estimate 12-step ahead point forecasts for panel data using an expanding window (window size to start with =60). My workfile ranges from 1987:01 to 2014:05 (IS: 1987:01 to 2009:12 and OOS: 2010:01 to 2014:05). I want to analyse analyst estimates for stocks and they publish 12-months ahead forecasts so my idea is to forecast 1992:12 using 1987:01 to 1991:12, then forecast 1993:01 using 1987:01 to 1992:01 and so on.
I found a code in this forum and tried to adjust it to my data, however, I get an error message saying "Date does not uniquely define observation in panels in "IF@DTOO ("LAST") < @DTOO("") THEN". In the eviews Command and Programming Reference I found that using DTOO in panel data will always cause an error message. How can I then specify the observation that my code will work?
Code: Select all
'Set window size
!window=60
'set step size
!step=12
'get size of workfile
!length=@obsrange
'declare equation for estimation
equation eq_dy_eps
'calculate number of rolls
!nrolls=@floor((!length-!window)/!step)
'matrix to store coefficient estimates
matrix (3,!nrolls) coefmat ' where 3 is the number of coefficients
'series to store forecast estimates
series fcast
'redundant series for catching start and end points
%start="@first" 'otod(@ifirst(ser))
%end="@last" 'otod(@ilast(ser))
'variable keeping track of how many rolls we've done
!j=0
'move sample !step obs at a time
for !j=1 to !length-!window+1-!step step !step
!j=!j+1
'set sample for estimation period
%first=@otod(@dtoo(%start)+!j-1)
%last=@otod(@dtoo(%start)+!j+!window-2)
smpl {%first} {%last}
'estimate equation - where the equation is LS
eq_dy_eps.ls dl_ri c dy_act(-12) eps_act(-12)
'store coefficients
colplace(coefmat,eq_dy_eps.@coefs,!j)
'12-period-ahead forecast
%12pers=@otod(@dtoo(%start)+!j+!window-1) 'start point %12pere=@otod(@dtoo(%start)+!j+!window+2) 'end point: %12pere-%12pers+1=12
if @dtoo (%end) < @dtoo(%12pere) then 'check whether the forecast end point is greater than the workfile end point
exitloop
endif
'set smpl for foreasting period
smpl {%12pers} {%12pere}
'forecast with command *forecast* (see also *fit*)
eq_dy_eps.forecast(f=na) yf
'set smpl to obtain the 12th period observation
smpl {%12pere} {%12pere}
'store forecasts
fcast=fceq
next
smpl @all
show coefmat
show fcast.line
d(noerr) ser