组间系数差异的检验方法- 引入交叉项(Chow 检验)
构建一个带交互效应的回归,如果交互项不显著,则说明两个回归系数是无显著区别的。
When two slope coefficients are different, a one-unit change in a predictor is associated with different mean changes in the response. In the scatterplot below, it appears that a one-unit increase in Input is associated with a greater increase in Output in Condition B than in Condition A. We can see that the slopes look different, but we want to be sure this difference is statistically significant.
我们需要确定 Input 的系数是否取决于 Condition。在统计学中,当我们说一个变量的影响取决于另一个变量时,这是一种交互效应。我们需要做的就是包含 Input*Condition 的交互项
#以普通线性回归为例,分别拟合了 fine 和 coarse 的降水和植被指数ndvi
fit_fine <- lm(NDVI_2018~pre_2018, data = sd2018_fine)
summary(fit_fine)
fit_coarse <- lm(NDVI_2018~pre_2018, data = sd2018_coarse)
summary(fit_coarse)
#比较交互效应,在回归中加入 Sepal.Length*Species 的交互项
fit_all <- lm(NDVI_2018~pre_2018*Soil, data = sd2018_all)
summary(fit_all)
https://blog.minitab.com/en/adventures-in-statistics-2/how-to-compare-regression-lines-between-different-models
https://www.zhihu.com/question/279103129/answer/2036840516
https://www.stata.com/support/faqs/statistics/computing-chow-statistic/index.html
https://mp.weixin.qq.com/s/CzBfLwubLcSvkmhyPLHZ0A