Residuals in MA model

General
1

What is the application when residuals are random, having more +ve residuals then -ve, and residuals being autocorrelated as stated below :
From the first and second plot above, you can see that the residuals are random and are more positive than negative. Hence the model made lower predictions in most cases.

From the second plot above, you can see that lag 16 lies outside the blue region. Hence, this indicates the presence of autocorrelation in the residuals.

Course Name: Financial Time Series Analysis for Trading, Section No: 15, Unit No: 6, Unit type: Notebook

2

Hi,

If your residuals are autocorrelated and mostly positive, it means your model is consistently underestimating returns and missing key patterns in the data. Since past residuals influence future ones, it’s a sign that important market behaviors like momentum or mean reversion aren’t being properly accounted for. So basically, there’s still valuable information left on the table that the model isn’t capturing.

3

Thanks for the reply