In case you're wondering what the mean absolute error and R-squared score mean, here is a quick overview. MAE quantifies the average magnitude of errors in a set of predictions, providing a direct measure of how far off the predictions are from the actual values. R-squared on the other hand, also known as the coefficient of determination, evaluates the goodness of fit of a statistical model. In simpler terms, it indicates how well the model fits the observed data. For more explanation on these terms, you can watch this video. Mean Absolute Error (MAE) is exactly what it sounds like - it measures how far off your predictions are from the actual values, on average. Think of it as the typical prediction error in the same units as your target variable. Lower is better, with zero being perfect predictions.
R-squared (R²) tells a different story - it shows how much of the variation in your target variable your model explains. It ranges from 0 to 1, where 1 means your model perfectly explains all the variation, and 0 means it's no better than just guessing the average. In real-world terms, our model with an R² of 0.563 on test data means it's explaining about 56.3% of the variation in yearly spending.
These two metrics together give you a pretty good picture of how well your model is performing. The MAE tells you about the size of your errors, while R² tells you how much of the pattern in the data your model has captured. For serious model evaluation, you'd want to look at both rather than just one or the other.