Evaluating Time-Series Model Performance

Easy Time-series Analysis Model Evaluation

A financial analyst is evaluating a time-series model that predicts monthly sales for a retail company. The model has been assessed using several statistics, and the analyst is particularly interested in understanding the effectiveness of the forecasting method employed.

The two simplest measures used for model evaluation in time-series analysis are Mean Absolute Error (MAE) and Root Mean Square Error (RMSE). Given the following outputs from the model: MAE = 100 and RMSE = 150, the analyst is trying to decide how to interpret these values in the context of model performance.

The model performs adequately as the RMSE is higher than the MAE.

The difference between MAE and RMSE indicates that the model has significant prediction errors.

Both MAE and RMSE are acceptable metrics for assessing model fit regardless of their values.

Hint

Submitted12.8K

Correct10.0K

% Correct78%