Computing the Correlation Matrix
The method can be used to determine whether a relationship exists between two time series. The method calculates the correlation coefficients for several time series in each case in pairs and displays them in a table. The degree of correlation is indicated by colors: The scale ranges from green (full correlation) to red (no correlation).
The values can range from -1 (completely negative linear correlation) to +1 (completely positive correlation). If the value is 0, there is no correlation between the values of the time series. If the correlation coefficient is greater than 0.5, there is a positive correlation.
The following figure illustrates schematically how the correlation matrix is determined.
Applying the Method
To start the method, select Methods → Analyze time series → Compute correlation matrix from the menu bar or the context menu of the time series in the Project Explorer (see figure below).
Use the Select Time Series button in the window that opens (see the figure below) to select the time series to be edited. A multiple selection is possible.
After clicking Apply, the result is shown in a table as in the following figure.
A double click on a cell colors this cell light blue and provides detailed information below (see following figure)
Example
The method can be used to determine the dependence of a heat demand on weather data (see following figure).
