The measure for predictive performance was calculated as the mean prediction error of the individual predicted minus observed HbA1c at end-of-trial for each arm.
The mean prediction error, R, proportion predicted outside the valid range, clinical interpretation of coefficients, model fit and estimation of Quality Adjusted Life Years (QALY) are reported and compared.
Thus, it is expected that the values in column 2 are numerically lower than the ones in column 3. Hbatc, at end-of-trial, was predicted with a mean prediction error of 0.06% (ranging from 0.0 to 0.13%) and 0.14% (ranging from 0.01 to 0.24%), when using all data (full model) and 12-week data only, respectively.
The data were analysed to compare the mean prediction error and the accuracy of predictability of intraocular lens power calculation between SRK II and Pediatric IOL Calculator.
This study is designed to compare the mean prediction error and the accuracy of predictability of the intraocular lens power calculation in pediatric patients after cataract surgery with primary implantation of intraocular lens using SRK II versus Pediatric IOL Calculator for pediatric IOL calculation.
The performance of each method was evaluated based on its ability to predict mean HbA1c at 26/28 weeks in each arm (by mean prediction error and RMSE) and on its ability to predict trial outcome expressed as the difference in treatment effect vs. comparator.
Fivefold cross-validation with 20% of the responses predicted in each run achieved a mean prediction error of 0.671, very close to the residual SD of 0.669.
The first was the mean prediction error (MPE), which measures the bias and predictive accuracy, and the second was the mean absolute prediction error (MAPE): MPE=frac{1}{n}{displaystyle sum_{i=1}^nleft {y}_i-{widehat{y}}_iright)} MAPE=frac{1}{n}{displaystyle sum_{i=1}^nleft|{y}_i-{widehat{y}}_ikern0.1em right|}.
The model was able to predict HbA1c at end-of-trial (24 28 weeks) with a mean prediction error of 0.14% ranging from 0.01 to 0.24% across the different treatment arms.
Both the single tree and tree cluster models were statistically similar and a combined model to predict average stem DBH yielded R2 = 0.71 with a mean prediction error (average DBH per stem) of ±13 cm within the range of 0.28 0.84 m.
The predictive performance of each model was evaluated using the model prediction error (PE), mean prediction error (MPE), mean absolute prediction error (MAPE), prediction-corrected visual predictive check (pcVPC), and normalized prediction distribution errors (NPDE).
