SSE(和方差.误差平方和):The sum of squares due to errorMSE(均方差.方差):Mean squared errorRMSE(均方根.标准差):Root mean squared errorR-square(确定系数):Coefficient of determinationAdjusted R-square:Degree-of-freedom adjusted coefficient of determination 下面我对以上几个名词进行详细的解释下,相
原文地址 ?传送门 对于回归预测结果,通常会有平均绝对误差.平均绝对百分比误差.均方误差等多个指标进行评价.这里,我们先介绍最常用的3个: 平均绝对误差(MAE) 就是绝对误差的平均值,它的计算公式如下: M A E ( y , y ^ ) = 1 n ( ∑ i = 1 n ∣ y − y ^ ∣ ) MAE(y,\hat{y}) = \frac{1}{n}(\sum_{i = 1}^{n}\left | y - \hat{y} \right |) MAE(y,y^)=n1(i=1∑n∣
TSS: Total Sum of Squares(总离差平方和) --- 因变量的方差 RSS: Residual Sum of Squares (残差平方和) --- 由误差导致的真实值和估计值之间的偏差平方和(Sum Of Squares Due To Error) ESS: Explained Sum of Squares (回归平方和) --- 被模型解释的方差(Sum Of Squares Due To Regression) TSS=RSS+ESS R2: Coefficien