Flowchart for Choosing a Regression Metric

I always wanted to create something like this for my students. This is straight out of chatGPT. I hope you find it useful - I think it is accurate but it may not be!.

Steps for the Flowchart:

1. Start
2. Is the regression problem linear or non-linear?Linear: Proceed to the next decision.Non-linear: Consider metrics like Mean Absolute Error (MAE), Root Mean Squared Error (RMSE), or Mean Squared Logarithmic Error (MSLE).
3. Are outliers a concern?Yes: Consider using MAE or Huber Loss (which is robust to outliers).No: Proceed to the next decision.
4. Is interpretability important?Yes: Use MAE or Mean Squared Error (MSE), as they are straightforward and commonly understood.No.: Proceed to the next decision.
5. Is there a need to penalize larger errors more severely?Yes: Use MSE or RMSE. No. Use MAE.
6. End with the chosen metric.

Regression Metrics to Consider:

• Mean Absolute Error (MAE): Measures the average magnitude of the errors in a set of predictions, without considering their direction.
• Mean Squared Error (MSE): Measures the average of the squares of the errors—that is, the average squared difference between the estimated values and the actual value.
• Root Mean Squared Error (RMSE): The square root of the MSE, gives a measure of the average magnitude of error. Penalizes larger errors more than MAE.
• Mean Squared Logarithmic Error (MSLE): Useful when dealing with data that has a wide range of values.
• Huber Loss: Combines the best properties of MSE and MAE. It is less sensitive to outliers in data than MSE.