FORECAST OUT-OF-SAMPLE ....HELP!!!
Posted: Sun Sep 23, 2012 4:50 pm
HEY PEOPLE
I NEED SOME HELP WITH THIS TOPIC
I RECENTLY DOING A WORKFILE WITH TIME SERIES AND A SMALL SAMPLE 2001 2011
MACRO VARIABLES, ANNUAL DATA
ALL OK (HETER.., AUTOCORR, RAMSEY TEST, ETC.) THEN WAS TO THE FORECASTING
FIRST, I ADJUST THE SAMPLE IN THE WORKFILE (PROC STRUCTURE/RESIZE...) 2001 2015
THEN I WENT TO THE FORECASTING AND ADJUST THE FORECAST SAMPLE 2001 2015
BUT THE OUTPUT WINDOW ONLY DISPLAY GRAPH UNTIL 2011 (ADJUSTED SAMPLE)
THEN I RUN THE FORECASTING AGAIN BUT ONLY FOR 2012 2015
AND PROMPT THE MESSAGE: "UNABLE TO COMPUTE DUE TO MISSING VALUES"
THE BOX FOR "INSERT THE ACTUALS FOR OUT-OF-SAMPLE OBSERVATIONS" WAS CHECKED PRVIOUSLY
SO WHAT HAPPEND??? !!!
THE MANUAL SAY THAT:
Insert actuals for out-of-sample observations.
By default, EViews will fill the forecast
series with the values of the actual dependent variable for observations not in the
forecast sample. This feature is convenient if you wish to show the divergence of the
forecast from the actual values; for observations prior to the beginning of the forecast
sample, the two series will contain the same values, then they will diverge as the forecast
differs from the actuals. In some contexts, however, you may wish to have forecasted
values only for the observations in the forecast sample. If you uncheck this
option, EViews will fill the out-of-sample observations with missing values.
IF I FILL WITH THE LAST OBSERVATION FOR EVERY INDEPENDET VARIABLE
THE FORECAST VALUES WILL BE THE SAME TO THE LAST VALUE OF THE DEPENDENT VARIABLE OBSERVED.
HELP ME
PLEASE!!!!!
I NEED SOME HELP WITH THIS TOPIC
I RECENTLY DOING A WORKFILE WITH TIME SERIES AND A SMALL SAMPLE 2001 2011
MACRO VARIABLES, ANNUAL DATA
ALL OK (HETER.., AUTOCORR, RAMSEY TEST, ETC.) THEN WAS TO THE FORECASTING
FIRST, I ADJUST THE SAMPLE IN THE WORKFILE (PROC STRUCTURE/RESIZE...) 2001 2015
THEN I WENT TO THE FORECASTING AND ADJUST THE FORECAST SAMPLE 2001 2015
BUT THE OUTPUT WINDOW ONLY DISPLAY GRAPH UNTIL 2011 (ADJUSTED SAMPLE)
THEN I RUN THE FORECASTING AGAIN BUT ONLY FOR 2012 2015
AND PROMPT THE MESSAGE: "UNABLE TO COMPUTE DUE TO MISSING VALUES"
THE BOX FOR "INSERT THE ACTUALS FOR OUT-OF-SAMPLE OBSERVATIONS" WAS CHECKED PRVIOUSLY
SO WHAT HAPPEND??? !!!
THE MANUAL SAY THAT:
Insert actuals for out-of-sample observations.
By default, EViews will fill the forecast
series with the values of the actual dependent variable for observations not in the
forecast sample. This feature is convenient if you wish to show the divergence of the
forecast from the actual values; for observations prior to the beginning of the forecast
sample, the two series will contain the same values, then they will diverge as the forecast
differs from the actuals. In some contexts, however, you may wish to have forecasted
values only for the observations in the forecast sample. If you uncheck this
option, EViews will fill the out-of-sample observations with missing values.
IF I FILL WITH THE LAST OBSERVATION FOR EVERY INDEPENDET VARIABLE
THE FORECAST VALUES WILL BE THE SAME TO THE LAST VALUE OF THE DEPENDENT VARIABLE OBSERVED.
HELP ME
PLEASE!!!!!