. Author manuscript; available in PMC: 2018 May 11.

Published in final edited form as: IEEE Trans Biomed Eng. 2016 Oct 25;64(8):1761–1771. doi: 10.1109/TBME.2016.2611862

Table III.

Regression Model Error Rates on Optimal Data Subset

Feature	Baseline	Linear Model		Quadratic Logistic GLM

	MAE	Data Subset	MAE	Data Subset	MAE
ENT	4.724	Weekday	4.432***	Weekday	3.968***
NENT	4.724	Optimized	4.207***	Weekday	4.125***
LV	4.724	Optimized	4.276***	Optimized	4.205***
HS	4.724	Weekday	4.602***	Weekday	4.197***
TT	4.724	Weekend	4.530***	Weekend	4.520**
TD	4.724	Base	4.664**	Median	4.666
NC	4.724	Weekend	4.516**	Weekend	4.576*
DM	4.724	Optimized	4.262***	Optimized	4.215***
DMN	4.724	Optimized	4.629**	Optimized	4.618*
DMD	4.724	Optimized	4.491***	Optimized	4.503***
Combined	4.724	Ten Features	3.748***	14 Features	3.737**

Significance of fitted model from the baseline model indicated by asterisks: * < 0.05; ** < 0.01; *** < 0.001.

Feature abbreviations are given in Table II; MAE: mean absolute error.