Regression with Outlier

Bad data such as outliers, noise, or drift can affect regression results. It is preferable to remove the bad data, but it is challenging to remove all bad data. This exercise demonstrates how to use an l1-norm objective to minimize the effect of a few outliers in the data with Python Gekko.
Source code: apmonitor.com/do/index.php/Ma...

[@dr.alikhudhair9414]

Wonderful

[@HuyNguyen-bw4sv]

Thank you!

[@smarkelov]

There is also a set of robust liner estimators in the scikit-learn library for working with outline data points.

[@apm]

Thanks for that comment. Combining sklearn with lazypredict helps with the regression evaluation: apmonitor.com/dde/index.php/Main/BiomechanicRegression