In a regression model, you are evaluating the performance of your predictions using the mean absolute error (MAE). If you notice that the MAE is significantly lower than the root mean square error (RMSE) for the same model, what can you infer about the distribution of the errors in your predictions?

When MAE is much smaller than RMSE, it means the squared errors are inflating the average, so a few large mistakes are pulling RMSE up. In other words, most predictions are close to the true values, but a handful of predictions are far off, creating outliers. This tells you that the error distribution is skewed or has a heavy tail rather than being tight around zero. For example, if most errors are ±1 but a few are ±10, the MAE will stay near 1 while RMSE rises toward about 4, showing the influence of those large errors. Thus, a lower MAE compared to RMSE indicates a few large deviations in the predictions.