Low R2 value

[mahaja1]

I see my R2 for testing data to be 0.215. Should i still rely on the model? Also, what are the ways I can improve the accuracy of the model?

[akim-collab]

Hi! R^2 is recommended to look only at the training dataset. To view the quality of a model on a test dataset, you need to look at the standard quality metrics of regression models, for example, MAPE, MAE, RMSE. I also strongly recommend conducting statistical tests for the model residuals on the training dataset.