I have two model prediction results:
- Using ARIMA model
- Using Machine learning model where I used Random Forest Regressor
How do we compare these two? Is conventional time series modelling better or, are machine learning methods better?
$\begingroup$
$\endgroup$
2
-
$\begingroup$ Hi Athos, welcome to the community. Please consider upvoting or marking the answer as correct, if you find anything helpful. $\endgroup$– KritiJun 23 at 13:42
-
$\begingroup$ Better at what? $\endgroup$– user2974951Jun 28 at 5:09
Add a comment
|
1 Answer
$\begingroup$
$\endgroup$
You can validate the model and compare one model with other by calculating accuracy metrics such as - R-squared, root mean squared error, residual standard error, mean absolute error, Adjusted R-squared, AIC, AICc, BIC.
Currently, the most popular metrics for evaluating time series forecasting models are MAE, RMSE and AIC
