Proc ESM Seasonal Model Help

[adscottsoden]

Hi, Im trying to do a forecast on some data for my group and using the Proc ESM...I do a seasonal daily model because I want it to account for changes in day of week. However, I also want to account for the month and other trends. I have one variable i'm wanting to forecast...but Im not getting as close as I want....do you have any suggestions of techniques for improving the signficance of the projected values and minimizing the errors?

Proc Sort data = cwork.forecast_queue_elig_vdps_1;
by Channel tier;
run;

ODS Listing;
ODS RTF File = '\\wfsnccicnsf01\c_cds_users\a097529\My Documents\forecastqueueEligvdps1.rtf';
proc ESM data=cwork.forecast_queue_elig_vdps_1 out = forecastedqueue_elig_vdps Lead = 15 plot=forecasts PRINT = FORECASTS;
by Channel tier;
ID trandate interval = day accumulate=average;
forecast totalqueuedin/model= seasonal;

run;
ODS rtf Close;

Thank you for your help!!!!

Allie

[udo_sas]

Allie -

Would it be possible to share your data as well?

Thanks,

Udo

[adscottsoden]

how do I send data? is it safe enough to attach to here?

[udo_sas]

Hello -

Yes - if you are willing to share your data just attach it to a post.

Thanks,

Udo

[ets_kps]

Hi Allie,

In addition to Udo's request I thought i'd push you in a certain direction.

The types of multiple seasonal/cycle models can be estimated in SAS using a UCM framework.  You can think of this as ESM-like but with regressors.  These regressors might be continous, such as a price, or dichotomous, such as a day.

You might want to experiment a bit with the combination of seasonal length as well as some dummies for month to take care of those cycles.  The dummy variable approach will require a good deal of data but as we don't know what you are working with, I'll just toss some ideas your way.

SAS/ETS(R) 12.1 User's Guide

Udo, might have more to add with some data.

Regards-Ken