You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I'm looking at outlier robust covariance estimators #3221, but I'm using a time series.
Can we add observation weights capturing "inlyingness" to the HAC aggregation of auto-covariances?
If we just use threshold or hard-rejection weights, then this would correspond to HAC with missing observations.
This assumes that observations are outliers, instead we might think of outliers in the correlation (i.e. think of stacked data as in VAR).
(I'm not sure how this works and no reference. Triggered because I used a cointegration test dataset for the robust covariance calculation, i.e. some variables in the data are integrated I(1))
The text was updated successfully, but these errors were encountered:
I'm looking at outlier robust covariance estimators #3221, but I'm using a time series.
Can we add observation weights capturing "inlyingness" to the HAC aggregation of auto-covariances?
If we just use threshold or hard-rejection weights, then this would correspond to HAC with missing observations.
This assumes that observations are outliers, instead we might think of outliers in the correlation (i.e. think of stacked data as in VAR).
(I'm not sure how this works and no reference. Triggered because I used a cointegration test dataset for the robust covariance calculation, i.e. some variables in the data are integrated I(1))
The text was updated successfully, but these errors were encountered: