CAS MA 575 Lecture Notes - Lecture 17: Dependent And Independent Variables

• School: Boston University
• Department: Mathematics & Statistics
• Course: CAS MA 575
• Professor: Ting Zhang

Document Summary

10. 1 variable selection based on hypothesis testing: forward algorithm. Consider all possible models with one explanatory variable. Choose the one with the smallest p-value and add it to the model if significant: repeat until none of the remaining variables is significant. (cid:885) (cid:1865)(cid:1867)(cid:1856)(cid:1857)(cid:1864)(cid:1871) (cid:1875)(cid:1872) (cid:1867)(cid:1873)(cid:1872) (cid:1872) (cid:1857) (cid:1866)(cid:1872)(cid:1857)(cid:1870)(cid:1855)(cid:1857)(cid:1868)(cid:1872): (cid:1851)~(cid:1876),(cid:2869) (cid:883) (cid:1851)~(cid:1876),(cid:2870) (cid:883) (cid:1851)~(cid:1876),(cid:2869) (cid:883) (cid:1851)~(cid:1876),(cid:2871)+(cid:1876),(cid:2869) (cid:883) (cid:1851)~(cid:1876),(cid:2871)+(cid:1876),(cid:2870) (cid:883) (cid:1876),(cid:2869) (cid:1867)(cid:1870) (cid:1876),(cid:2870) which one is significant (using t test) and add the one with smaller p-value. >summary(lm(y~x3-1)) (cid:1868) the most significant, include this. >summary(lm(y~x3+x1-1)) (cid:1868) the most significant, include this. >summary(lm(y~x3+x2+x1-1)) (cid:1868)=(cid:882). (cid:889)(cid:885)9 summary(lm(y~x1+x3))