Calculates the Variance Accounted For between two signals.


v = vaf(y,yhat)


The function vaf calculates the Variance Accounted For between two signals. The VAF between y and yhat for the i th component is defined as

$$ \mathrm{VAF}_i = \left(1 - \frac{\mathrm{var}(y_i -
\hat{y}_i)}{\mathrm{var}(y_i)} \right) \cdot 100 \% $$

The VAF of two signals that are the same is 100%. If they differ, the VAF will be lower. If y and yhat have multiple columns, the VAF is calculated for every column in y and yhat seperately.

The VAF is often used to verify the correctness of a model, by comparing the real output with the estimated output of the model.


y is the measured output y(k).

yhat is the estimated output yhat(k).


v is the Variance Accounted For as defined above.

Used By

This is a top-level function that is used directly by the user.