Skip to main content

Table 2 Comparisons of fit statistics

From: Mapping the EORTC QLQ-C30 to EQ-5D-3L in patients with breast cancer

  MAE RMSE AICa BICa QICa Meanb Absolute Difference
Linear regression 0.1203 0.1680 0.7103 0.0047
Response mapping 0.1191 0.1702 0.7057 0.0093
1 component ALDVMM 0.1196 0.1696 − 2302.11 − 2164.82 − 2289.01 0.7201 0.0051
2 component ALDVMM 0.1173 0.1684 − 2538.83 −2164.82 − 2515.96 0.7201 0.0051
3 component ALDVMM 0.1172 0.1677 − 2528.40 − 2154.39 − 2459.09 0.7171 0.0021
4 component ALDVMM 0.1173 0.1675 − 2673.51 − 2168.59 − 2629.49 0.7182 0.0032
  1. MAE Mean Absolute Error, RMSE Root Mean Squared Error, AIC Akaike Information Criteria, BIC Bayesian Information Criteria
  2. aNote that AIC, BIC and QIC cannot be compared between linear regression, response mapping and mixture models and so they are not included for the linear model or response mapping
  3. bmean observed = 0.7150