线性回归系数的置信区间应基于正态分布还是
让我们有一些线性模型,例如简单的方差分析: # data generation set.seed(1.234) Ng <- c(41, 37, 42) data <- rnorm(sum(Ng), mean = rep(c(-1, 0, 1), Ng), sd = 1) fact <- as.factor(rep(LETTERS[1:3], Ng)) m1 = lm(data ~ 0 + fact) summary(m1) 结果如下: Call: lm(formula = data ~ 0 + fact) Residuals: Min 1Q Median 3Q Max -2.30047 …