blockCorrelationTest
Tests the null hypothesis that two signals are uncorrelated
Extends from Modelica.Blocks.Interfaces.SI2SO (2 Single Input / 1 Single Output continuous control block).
Information
This block first calculates the Correlation
of its two input signals (with delta_t = 0) and performs a
SignificanceTest against the null-hypothesis
that the signals are uncorrelated. If the output of this block is lower
than a threshold of p < 0.05 or p < 0.01,
the two inputs must be assumed to be correlated. If the output is higher
than this treshold, the two inputs are probably not correlated.
For more information see models linked above.
This correlation test uses a Fisher transformation to shape the correlation
coefficient r to a more normally distributed number. More information
on this procedure can be found e.g. on Wikipedia:
https://en.wikipedia.org/wiki/Fisher_transformation.
Parameters
| Type | Name | Default | Description |
|---|
| Modelica.Units.SI.Time | t_eps | 0.001 | Start evaluating shortly after simulation start |
| Real | rho | 0.0 | Expected correlation of null hypothesis (=0) |
Contents
| Name | Description |
|---|
 FisherTransformationprotected | Transforms a correlation to be approximately normally distributed |
