Inspect Seasonality

This component calculates autocorrelation with Pearson Correlation for lagged copies of time series. Additionally, it produces an interactive view that displays the Autocorrelation Function (ACF) Plot, Partial Autocorrelation Function (PACF) Plot, and detects the first local maximum of correlation for sign of dominant seasonality.

If you encoutner errors please verify that
Preferances > KNIME > Python (labs) > Python environment configuration
is set to bundled

Options

Value Column
Column to apply (P)ACF analysis to. This column has to be a numeric column.
Seasonality Cut Off
The component will only consider correlations above the cut off values as seasonal trends. This value has to be between 0 and 1.
Max Lag
Maximum lag to use when checking for (partial) autocorrelation.
Step size of the moving ACF window
Size of steps to move when checking (partial) autocorrelation of different lags. This value should not exceed the value of Max Lag.

Input Ports

Icon
Data containing selected column for the Autocorrelation Function (ACF) and Partial Autocorrelation Function (PACF) analysis.

Output Ports

Icon
This table contains a list of detected local maximum, their corresponding correlation, and lag value.
Icon
This flow variable contains the lag value where the dominant seasonality might occur.

Nodes

Extensions

Links