Table 5.

A comparison of MLM model performance on the prediction of stress.

Model	Fixed effects					Goodness of fit and comparison*
	Estimate	SE	t	df	p-value	F	df1, df2	P-value
Baseline model
Intercept	0	0.12	0.00	105	> 0.999	282.3	290.4	–
EMA model
Intercept	0	0.10	0.00	104	> 0.999	11.09	1, 434	0.001a
Arousal	−0.39	0.11	−3.71	74	<0.001
GPS model
No predictors identified
Wearable data model
No predictors identified
Combined model: EMA and digital phenotyping model
Not applicable

^*For more details on the likelihood ratio test see chapter 5.3.1 in Van Buuren (66).

^aComparison against baseline.