Correlation Analysis Wizard: Analysis Options
Correlation Analysis Dialog: Analysis Options
The Correlation Analysis Wizard allows you to search for correlation between the data in one field and the previous or future values of one or more other fields. This is useful for determining if one field may be a good input for predicting another. The Analysis Options page allows you to specify the date range and refine the type of analysis.
Selecting a Date Range for Correlation Analysis
Correlation analysis is performed over a series of contiguous dates in the data series of the selected fields and the field being analyzed. By default, the correlation analysis wizard will use all of the dates for which there is data in these data series. To select a subset of the available data, adjust the date or size values.
¨ Start of Range / End of Range
These values specify the start and end of the data range. Use the date controls to increase or decrease these values. As these values are updated, the size will be updated automatically.
& For help using the date controls, see the help for Date Selection Controls.
Ä Note: The data available for the correlation will be limited by the date ranges available for all of the fields in the data being analyzed. If you would like to analyze a larger date range than is currently available, try removing the smaller data series. You can quickly identify the start and end dates of all data series by adding the start and end dates as display fields in the portfolio view.
¨ Size
This value specifies the number of samples, days, or other length of data in the selected range. If this value is changed, the range will be updated to match the new size. If the End of Range is set to the last date, the Start of Range will be adjusted; otherwise, the End of Range will be adjusted.
Ä Note: Size in samples is not available when analyzing group fields since the number of samples may be different for the various data series in the group.
¨ Portion to Analyze
This value specifies the percentage of the specified range that should be analyzed for correlation. This allows you to specify the entire range that would be used for a prediction and exclude the portion that would be used for the test set. This value defaults to the default size for the accuracy testing range.
If you have an existing field that you would like to copy the date range from, press the Copy Optimization Range from a Field… button. This will display the Copy Optimization Range From Field Dialog.
Selecting Options to Use During Correlation Analysis
The following options can be set for the correlation analysis.
p Field Being Analyzed
This value displays the name of the field being analyzed. It is repeated from the title of the dialog and can not be changed.
Ä Note: To analyze a different field, exit the correlation analysis for this field. Then, select the field you would like to analyze.
¨ Type of Analysis
This value displays the type that was selected on the Analysis Type page. It is repeated here so that the type of previous correlation analysis reports can be changed.
¨ Field Preprocessing
This value indicates the type of preprocessing that should be applied to field being analyzed.
· Percent Change
This option uses the percent change in the value of the field. It is useful for analyzing fields that do not have a constant range, such a price data.
· Change
This option uses the change in the value of the field. It is useful for analyzing fields that have both positive and negative values and those that include zero.
· None
This option uses the actual value of the field. It is useful for analyzing fields that have a consistent range, such as entry/exit signals and some financial indicators.
Ä Note: The Prediction Samples in Advance value determines the number of samples included in the change or percent change preprocessing. For example, five samples in advance uses the change from the current value to the value five days in the future. Because of this, preprocessing is not available when Samples in Advance is set to 0.
¨ Prediction Samples in Advance
This value indicates the number of samples in advance a prediction would be made for based on this analysis. It is used in addition to the lags to determine the number of samples between this field and the fields being analyzed. It is also modifies the Percent Change and Change field preprocessing.
Ä Note: This value is restricted to 1 when analyzing this field as a potential inputs to other fields.
¨ Input Lags to Analyze
These values indicate the potential lags to test for the field or fields being analyzed. This simulates the use of memory or lagged inputs in a prediction. Using a minimum lag of 0 includes the current value of each potential input being tested.
What Do I Do Next?
If you are creating a new correlation analysis, press the Finish button to start the correlation analysis. After it has completed, you will be returned to the Modify Field Dialog: Correlation Analysis page to view the results. If you would like to exit the Wizard without performing any analysis, press the Cancel button. If you would like to change the fields being analyzed, press the Back button to return to the Analysis Fields page.
If you are modifying a previous correlation analysis, press the OK button to save your changes or select another property page. If you would like to exit from this dialog without changing the previous correlation analysis, press the Cancel button.
Additional Property Pages
The Correlation Analysis Wizard is displayed as a dialog with property pages when previous correlation analysis options are being modified. In addition to this page, the Analysis Fields page is also available.
& For help using property pages, see the help for Property Pages.
How Did I Get Here?
The Analysis Fields page appears after the Analysis Fields page in the Correlation Analysis Wizard. It is also available as a property page when you press the Change Analysis… button from the Modify Field Dialog: Correlation Analysis page.