Creating the Model

  1. On the Time Series node, choose the Fields tab. In the Fields list, select all 5 of the markets and copy them to both the Targets and Candidate inputs lists. In addition, select and copy the Total field to the Targets list.
  2. Choose the Build Options tab and, on the General pane, ensure the Expert Modeler Method is selected using all default settings. Doing so enables the Expert Modeler to decide the most appropriate model to use for each time series. Click Run.
    Figure 1. Choosing the Expert Modeler for Time Series
    Choosing the Expert Modeler for Time Series
  3. Attach the Time Series model nugget to the Time Series node.
  4. Attach a Table node to the Time Series model nugget and click Run.
Figure 2. Sample stream to show Time Series modeling
Sample stream to show Time Series modeling

There are now three new rows (61 through 63) appended to the original data. These are the rows for the forecast period, in this case January to March 2004.

Several new columns are also present now; the $TS- columns are added by the Time Series node. The columns indicate the following for each row (that is, for each interval in the time series data):

Column Description
$TS-colname The generated model data for each column of the original data.
$TSLCI-colname The lower confidence interval value for each column of the generated model data.
$TSUCI-colname The upper confidence interval value for each column of the generated model data.
$TS-Total The total of the $TS-colname values for this row.
$TSLCI-Total The total of the $TSLCI-colname values for this row.
$TSUCI-Total The total of the $TSUCI-colname values for this row.

The most significant columns for the forecast operation are the $TS-Market_n, $TSLCI-Market_n, and $TSUCI-Market_n columns. In particular, these columns in rows 61 through 63 contain the user subscription forecast data and confidence intervals for each of the local markets.

Next