<?xml version="1.0" encoding="UTF-8"?>
<rss xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#" xmlns:taxo="http://purl.org/rss/1.0/modules/taxonomy/" version="2.0">
  <channel>
    <title>topic Forecast Studio:  Odd Model Selection in SAS Forecasting and Econometrics</title>
    <link>https://communities.sas.com/t5/SAS-Forecasting-and-Econometrics/Forecast-Studio-Odd-Model-Selection/m-p/14942#M50</link>
    <description>Hello,&lt;BR /&gt;
&lt;BR /&gt;
I am using Forecast Studio to create monthly forecasts.  I am encountering a problem where the forecast model selection creates an exponential increase for the last three months when the actuals show a basically constant history.  Graphically it looks like a 'hockey stick'.&lt;BR /&gt;
&lt;BR /&gt;
I then drill into that specific series and select another forecast model that better reflects the data.  It may have a greater MAPE but it makes much more sense for the long term projection.&lt;BR /&gt;
&lt;BR /&gt;
Has anyone ever encountered this situation?  If so, is there a programmatic or better way to deal with this?  My datasets are quite large with many levels of hierarchy.&lt;BR /&gt;
&lt;BR /&gt;
Thank you in advance.</description>
    <pubDate>Tue, 04 May 2010 15:40:53 GMT</pubDate>
    <dc:creator>Sonya_</dc:creator>
    <dc:date>2010-05-04T15:40:53Z</dc:date>
    <item>
      <title>Forecast Studio:  Odd Model Selection</title>
      <link>https://communities.sas.com/t5/SAS-Forecasting-and-Econometrics/Forecast-Studio-Odd-Model-Selection/m-p/14942#M50</link>
      <description>Hello,&lt;BR /&gt;
&lt;BR /&gt;
I am using Forecast Studio to create monthly forecasts.  I am encountering a problem where the forecast model selection creates an exponential increase for the last three months when the actuals show a basically constant history.  Graphically it looks like a 'hockey stick'.&lt;BR /&gt;
&lt;BR /&gt;
I then drill into that specific series and select another forecast model that better reflects the data.  It may have a greater MAPE but it makes much more sense for the long term projection.&lt;BR /&gt;
&lt;BR /&gt;
Has anyone ever encountered this situation?  If so, is there a programmatic or better way to deal with this?  My datasets are quite large with many levels of hierarchy.&lt;BR /&gt;
&lt;BR /&gt;
Thank you in advance.</description>
      <pubDate>Tue, 04 May 2010 15:40:53 GMT</pubDate>
      <guid>https://communities.sas.com/t5/SAS-Forecasting-and-Econometrics/Forecast-Studio-Odd-Model-Selection/m-p/14942#M50</guid>
      <dc:creator>Sonya_</dc:creator>
      <dc:date>2010-05-04T15:40:53Z</dc:date>
    </item>
    <item>
      <title>Re: Forecast Studio:  Odd Model Selection</title>
      <link>https://communities.sas.com/t5/SAS-Forecasting-and-Econometrics/Forecast-Studio-Odd-Model-Selection/m-p/14943#M51</link>
      <description>Hello -&lt;BR /&gt;
It seems to me that a trend model is picked by SAS Forecast Studio - which apparently seems to be not appropriate for your data. &lt;BR /&gt;
Here is what you might want to consider:&lt;BR /&gt;
a) use a hold-out sample to select the model (see: Project/Forecasting Settings/Model selection)&lt;BR /&gt;
b) use a selection statistic different to MAPE like SBC or AIC (see: Project/Forecasting Settings/Model selection)&lt;BR /&gt;
c) run "Level Model Forecast Quality Extreme Table" report from "Reports and Stored Processes" as a post-process and deal with the flagged models interactively.&lt;BR /&gt;
Thanks,&lt;BR /&gt;
Udo</description>
      <pubDate>Mon, 10 May 2010 09:21:38 GMT</pubDate>
      <guid>https://communities.sas.com/t5/SAS-Forecasting-and-Econometrics/Forecast-Studio-Odd-Model-Selection/m-p/14943#M51</guid>
      <dc:creator>udo_sas</dc:creator>
      <dc:date>2010-05-10T09:21:38Z</dc:date>
    </item>
  </channel>
</rss>

