R语言中的Theil-Sen回归分析

原文链接：http://tecdat.cn/?p=10080

Theil-Sen估计器是一种在社会科学中不常用 的简单线性回归估计器 。三个步骤：

在数据中所有点之间绘制一条线

计算每条线的斜率

中位数斜率是 回归斜率

用这种方法计算斜率非常可靠。当误差呈正态分布且没有异常值时，斜率与OLS非常相似。

有几种获取截距的方法。如果 关心回归中的截距，那么知道 软件在做什么是很合理的。

当我对异常值和异方差性有担忧时，请在上方针对Theil-Sen进行简单线性回归的评论 。

我进行了一次模拟，以了解Theil-Sen如何在异方差下与OLS比较。它是更有效的估计器。

library(simglm) library(ggplot2) library(dplyr) library(WRS) # Hetero nRep <- 100 n.s <- c(seq(50, 300, 50), 400, 550, 750, 1000) samp.dat <- sample((1:(nRep*length(n.s))), 25) lm.coefs.0 <- matrix(ncol = 3, nrow = nRep*length(n.s)) ts.coefs.0 <- matrix(ncol = 3, nrow = nRep*length(n.s)) lmt.coefs.0 <- matrix(ncol = 3, nrow = nRep*length(n.s)) dat.s <- list() ggplot(dat.frms.0, aes(x = age, y = sim_data)) + geom_point(shape = 1, size = .5) + geom_smooth(method = "lm", se = FALSE) + facet_wrap(~ random.sample, nrow = 5) + labs(x = "Predictor", y = "Outcome", title = "Random sample of 25 datasets from 15000 datasets for simulation", subtitle = "Heteroscedastic relationships")

need-to-insert-img

need-to-insert-img

ggplot(coefs.0, aes(x = n, colour = Estimator)) + geom_boxplot(  aes(ymin = q025, lower = q25, middle = q50, upper = q75, ymax = q975), data = summarise(   group_by(coefs.0, n, Estimator), q025 = quantile(Slope, .025),   q25 = quantile(Slope, .25), q50 = quantile(Slope, .5),   q75 = quantile(Slope, .75), q975 = quantile(Slope, .975)), stat = "identity") + geom_hline(yintercept = 2, linetype = 2) + scale_y_continuous(breaks = seq(1, 3, .05)) + labs(x = "Sample size", y = "Slope",   title = "Estimation of regression slope in simple linear regression under heteroscedasticity",   subtitle = "1500 replications - Population slope is 2",   caption = paste(    "Boxes are IQR, whiskers are middle 95% of slopes",    "Both estimators are unbiased in the long run, however, OLS has higher variability",    sep = "\n"   ))

need-to-insert-img

need-to-insert-img