suppressor_in_multiple_regression
Differences
This shows you the differences between two versions of the page.
| Both sides previous revisionPrevious revisionNext revision | Previous revision | ||
| suppressor_in_multiple_regression [2025/11/09 12:42] – [도식화] hkimscil | suppressor_in_multiple_regression [2025/11/11 01:09] (current) – hkimscil | ||
|---|---|---|---|
| Line 477: | Line 477: | ||
| summary(lm.y.x2)$r.squared = 0 | summary(lm.y.x2)$r.squared = 0 | ||
| lm.y.x1 <- lm(y ~ x1) | lm.y.x1 <- lm(y ~ x1) | ||
| - | summary(lm.y.x1)$r.squared = a | + | summary(lm.y.x1)$r.squared = a, p > .05 (may be not significant) |
| + | # 그러나 아래처럼 x2가 포함됨으로써 | ||
| + | # x1 내의 x2와 관련된 부분이 제거되어 | ||
| + | # 분석됨으로 significant하게 됨 | ||
| lm.y.x1x2 <- lm(y ~ x1 + x2) | lm.y.x1x2 <- lm(y ~ x1 + x2) | ||
| - | summary(lm.y.x1)$r.squared = a # but, with x1 without overlapped area with x2 as explaining part | + | summary(lm.y.x1)$r.squared = a |
| </ | </ | ||
| Line 504: | Line 506: | ||
| > lm.x1.x2 <- lm(Price~CompPrice, | > lm.x1.x2 <- lm(Price~CompPrice, | ||
| > res.x1x2 <- lm.x1.x2$residuals | > res.x1x2 <- lm.x1.x2$residuals | ||
| - | > lm.y.resx1x2 | + | > lm.y.onlyx1<- lm(Sales~res.x1x2) |
| > summary(lm.y.onlyx1)$r.squared | > summary(lm.y.onlyx1)$r.squared | ||
| [1] 0.3537271 | [1] 0.3537271 | ||
| Line 517: | Line 519: | ||
| [1] 0.1979812 | [1] 0.1979812 | ||
| > | > | ||
| - | + | > summary(lm.y.x1)$r.squared | |
| + | [1] 0.1979812 | ||
| + | > pcor.test(Sales, | ||
| + | estimate | ||
| + | 1 -0.5959744 9.95161e-40 -14.78787 400 1 pearson | ||
| + | > pcor.test(Sales, | ||
| + | [1] 0.3551855 | ||
| + | > summary(lm.y.x2)$r.squared | ||
| + | [1] 0.004106084 | ||
| + | > | ||
| </ | </ | ||
suppressor_in_multiple_regression.1762692155.txt.gz · Last modified: by hkimscil