Summary statistics for the fall risk models using data from the Markov chains.

R2 gives the adjusted R2 value for the fitted model (Eq 12). The root mean squared error (RMSE), mean absolute error (MAE), mean absolute percentage error (MAPE), relative average absolute error (RAAE), and maximum absolute error (Max Abs Err) are over all training conditions. The % in Tolerance indicates the percentage of these trials in which the predicted value was within ln(MFPT)±3.3. The slope, intercept, and are from Eq 15. Values of 1, 0, and 1, respectively, would indicate a perfect fit.

调整决定系数（adjusted R²）为拟合模型（式12）的对应取值。均方根误差（RMSE, root mean squared error）、平均绝对误差（MAE, mean absolute error）、平均绝对百分比误差（MAPE, mean absolute percentage error）、相对平均绝对误差（RAAE, relative average absolute error）以及最大绝对误差（Max Abs Err, maximum absolute error）均基于全部训练条件计算得到。容差占比（"% in Tolerance"）指的是预测值落在自然对数形式的平均首通时间（MFPT, mean first passage time）±3.3范围内的试验所占百分比。斜率、截距与[原文缺失项]均来自式15。若该三项分别取值为1、0与1，则代表模型拟合效果完美。