IBM Support

Time Series Modeler table showing forecasts is limited to 200 time points

Troubleshooting


Problem

I'm fitting time series models in SPSS using the Time Series Modeler (TSMODEL procedure). I want to get the table of forecasts to show the forecasts and confidence interval bounds for all of the forecasts, but I get a warning message instead: The forecast table has reached its maximum size. Only the first 200 forecasts are shown. Can I get more than 200 time points shown in this table?

Resolving The Problem

The limit of 200 time points in the table of forecasts is a hard-coded limit and there is currently no way to change it. An enhancement request has been filed with SPSS Development asking that options for producing tables of forecasts with more than 200 time points be added to the TSMODEL procedure. For the time being, you can produce a table of forecasts of any length by saving the forecasts and confidence interval bounds and using the SUMMARIZE procedure (Analyze>Reports>Case Summaries in the menus). For example, the following commands fit models to the variables v1 and v2 using the Expert Modeler, save the forecasts and confidence interval bounds, and produce a table of forecasts for cases beyond the observed data (assuming that these cases have missing values for both v1 and v2):

PREDICT THRU END.
TSMODEL
/SAVE PREDICTED(Forecast) LCL UCL
/MODEL DEPENDENT=v1 v2.

TEMPORARY.
SELECT IF MISSING(v1) AND MISSING(v2).

SUMMARIZE
/TABLES=DATE_ Forecast_v1_Model_1 LCL_v1_Model_1 UCL_v1_Model_1 Forecast_v2_Model_2 LCL_v2_Model_2 UCL_v2_Model_2
/FORMAT=LIST NOCASENUM NOTOTAL
/TITLE='Forecasts'
/MISSING=VARIABLE
/CELLS=NONE

[{"Product":{"code":"SSLVMB","label":"IBM SPSS Statistics"},"Business Unit":{"code":"BU059","label":"IBM Software w\/o TPS"},"Component":"Not Applicable","Platform":[{"code":"PF025","label":"Platform Independent"}],"Version":"Not Applicable","Edition":"","Line of Business":{"code":"LOB10","label":"Data and AI"}}]

Historical Number

81021

Document Information

Modified date:
16 April 2020

UID

swg21482748