. 2018 Jun 13;4(4):160–166. doi: 10.1192/bjo.2018.21

Table 3.

Comparison of model performance based on English value set for the EQ-5D-5L

Model	DASS-21						K-10
	Full sample estimation			Cross-validation			Full sample estimation			Cross-validation
	adj. r²	NMAE	NRMSE	adj. r²	NMAE	NRMSE	adj. r²	NMAE	NRMSE	adj. r²	NMAE	NRMSE
OLS	0.3320	0.1145	0.1539	0.3343	0.1146	0.1552	0.3288	0.1135	0.1543	0.3276	0.1147	0.1545
GLM	0.3324	0.1139	0.1541	0.3293	0.1135	0.1541	0.3285	0.1130	0.1544	0.3208	0.1126	0.1543
Beta binomial	0.3380	0.1159	0.1536	0.3334	0.1157	0.1552	0.3345	0.1150	0.1541	0.3336	0.1155	0.1560
FRM	0.3387	0.1135	0.1532	0.3309	0.1136	0.1531	0.3345	0.1125	0.1536	0.3324	0.1124	0.1541
MM-estimation	0.3318	0.1121	0.1575	0.3295	0.1121	0.1565	0.3287	0.1111	0.1574	0.3244	0.1108	0.1574
CLAD	0.3306	0.1121	0.1567	0.3262	0.1124	0.1566	0.3288	0.1111	0.1577	0.3283	0.1118	0.1607

Note. The best results are in bold type.

adj. r², square of correlation coefficient between predicted and observed EQ-5D-5L, penalised for number of predictors; CLAD, censored least absolute deviation; DASS-21, Depression Anxiety Stress Scales; FRM, fractional regression model; GLM, generalised linear model; K-10, Kessler Psychological Distress Scale; NMAE, normalised mean absolute error; NRMSE, normalised root mean square error; OLS, ordinary least squares regression.