Table 6.

Results of the leave-all-out time-series cross-validation and leave-one-out time-series cross-validation of the hierarchical Bayesian linear regression model, commonly used machine learning models, and the baseline model.

Model	Leave-all-out		Leave-one-out
	R2	RMSEa	R2	RMSE
Baseline modelb	0.338	4.547	−0.074	5.802
LASSO regression	0.458	4.114	0.144	5.178
XGBoost regression	0.464	4.092	0.346	4.523
Hierarchical Bayesian linear (second-order statistical features)	0.481	4.026	0.353	4.501
Hierarchical Bayesian linear (all Bluetooth features)	0.526	3.891	0.387	4.426

^aRMSE: root mean squared error.

^bThe baseline model is the hierarchical Bayesian linear regression model with only the last observed 8-item Patient Health Questionnaire score and demographics as predictors.