Weighted and Generalized Least
相逢一醉为前缘
这个作者很懒,什么都没留下…
展开
-
1 Weighted Least Squares
1. Focusing accuracyDiscounting imprecisionOften however the magnitude of the noise is not constant, and thedata are heteroskedastic.Sampling bias. In many situations, our data comes from a survey, an...转载 2018-05-31 15:24:55 · 3982 阅读 · 0 评论 -
2 Heteroskedasticity
Suppose the noise variance is itself variable.Figure 2: Scatter-plot of n = 150 data points from the above model. (Here X isGaussian with mean 0 and variance 9.) Grey: True regression line. Dashed: or...转载 2018-05-31 16:02:03 · 600 阅读 · 0 评论 -
3 The Gauss-Markov Theorem
Like any optimality result, it is crucial to lay out carefully the range of possiblealternatives, and the criterion by which those alternatives will be compared. Theclassical optimality result for est...转载 2018-05-31 18:47:50 · 863 阅读 · 0 评论 -
4 Finding the Variance and Weights
mobility <- read.csv("http://www.stat.cmu.edu/~cshalizi/mreg/15/lectures/24--25/mobility2.csv") plot(Mobility ~ Population, data=mobility, log="x", ylim=c(0,0.5)) mobility$MobSE &l...转载 2018-05-31 21:52:40 · 204 阅读 · 0 评论