Zhang and Lu (2012) proposed five different methods to estimate bias in RF for regression. Their methods estimateresiduals and add these estimated residuals to the predicted values to correct bias. Our method is similar to theirs becausewe use the estimated residuals, but we go further. We will fit a linear model with estimated residuals as the response variable(Y) and predicted values as the explanatory variable (X). Then, we will rotate this fitted line to the horizontal line or findthe best rotation angle to minimize bias.The RF package in R (Liaw & Wiener, 2009) offers a bias correction method using a simple linear regression (SLR). It fitsa SLR with observed values as the response variable (Y) and predicted values as the explanatory variable (X). One thingto note is that this built-in bias correction uses the fitted values from out-of-bag samples. We will later show that usinga bias correction with all data to fit a SLR offers better performance than a bias correction only with out-of-bag samplesin real data applications. The predicted values from out-of-bag samples can be computed as follows. Suppose we haveB bootstrap samples. Basically, tree bagging method takes an average of the predicted values from B trees. Suppose wehave to compute the predicted values for the first observation. Since they are bootstrap samples, there can be samplesthat the first observations are not included. Let us say that K(predicted values for the first observation from the out-of-bag samples are the average of the predicted values from these K
samples.
This paper is organized as follows. In Section 2, we introduce Zhang and Lu’s method (Zhang & Lu, 2012) and our two
methods. In Section 3, we compare the performances of six methods (original RF, built-in bias correction, bias correction
with SLR, Zhang and Lu’s first method, residual rotation, best angle rotation) with simulation data and five different real
data sets from the UCI data repository (Bache & Lichman, 2014). We give the concluding remarks in Section 4.
การแปล กรุณารอสักครู่..
