Skip to main content
. 2021 Mar 24;11:6713. doi: 10.1038/s41598-021-85987-9

Figure 2.

Figure 2

Comparison of models at the national level. (A) Root Mean Squared Error (RMSE) of fitted values, (B) model R-squared and (C) RMSE of out-of-sample predictions. The models are Support Vector Machines Regression (SVM), Random Forest Regression, Multivariable linear regression, and ARIMA with [i.e., ARIMA (covariates)] and without (i.e., ARIMA) search data as explanatory variables.