statsmodels.regression.recursive_ls.RecursiveLSResults.cusum_squares¶

RecursiveLSResults.cusum_squares¶

Cumulative sum of squares of standardized recursive residuals statistics

Returns:¶

cusum_squaresarray_like

An array of length nobs - k_exog holding the CUSUM of squares statistics.

Notes

The CUSUM of squares statistic takes the form:

$$s_t=\left(\sum _{j=k+1}^t{w}_j^2\right)/\left(\sum _{j=k+1}^T{w}_j^2\right)$$

where $w_j$ is the recursive residual at time $j$.

Excludes the first k_exog datapoints.

References