Closed ancapdev closed 2 years ago
Thank you. This will be fixed in 2.2.1, see #103
I still wouldn't put much trust in the estimation results in this particular case, because you're estimating 10 parameters from 8 data points (after losing 2 observations due to the lags), meaning the parameters are not identified (though I assume that in your real use case, the data set is larger).
The real data is much longer and it's sliding over a time series. I expect to have very low confidence at the start of the process. I had initially set my minimum window to 10 samples arbitrarily for testing and came upon this.
Thanks for the quick fix!
On version 2.2.0.
The following throws when trying to display the model (in a notebook)
Outputs:
Digging a little deeper,
volatilities(am)
returns an empty vector.Tracking further, from
volatilities()
:Returns
-Inf
and doesn't fill theht
vector.Not sure where to go from there. Random changes to the input make it work or not work and seems quite brittle.