Transcription of JMP による対話的パーティショニング
1 1 JMP SAS Institute Japan JMP 2009 5 JMP Q&A JMP Q&A Q1. A1. Chi-squared Automatic Interaction Detector(CHAID) Kass and Hawkins1 "Recursive Partitioning" JMP Partition( ) 2 Q2. JMP ? A2. Q3.
2 CART CHAID A3. JMP CART CHAID Q4. A4. JMP 8 [K ] [ ] [ ] R2 R2 ( 1) JMP 8 1 Hawkins, and Kass, (1982), Automatic InteractionDetection, in Hawkins, , ed., Topics in Applied Multivariate Analysis,267-302, Cambridge Univ Press:Cambridge.
3 2 JMPer Cable Spring 2005 Issue 17 2 1 (JMP ) [ ] JMP JMP 2 [ ] 2 SS " SS" SS JMP 2 G^2 " G^2" G^2 JMP X,Y 2 " SS" " G^2" [ ] G^2 [ ] 3 [ ]
4 [ ] p = -log10( p ) p X p Monte Carlo Calibration of Distributions of Partition Statistics 3 2 2 2 G^2 JMP.
5 Jmp [Y, ] [X, ] 4 2 2 [ ] [Y, ] [X, ] 3 2 (-1)* = 2 G^2 = 3 , , 3 3 2 3 a. b. c. a,b,c 2 4 a,b,c 3 2 b 2 2 G^2 = 4 18 60 2 2 2 36 / 5 2 5 G^2 = 5 2 G^2 [ ] ( 6)
6 6 G^2 G^2 G^2 (= ) G^2(= ) G^2 7 7 6 [ ] p p 3 (1) 8 8 (*) G^2 SS < G^2 SS > G^2 SS 9 9 JMP.
7 Jmp [Y, ] [X, ] 7 * SS SS 2 , 2 2 [ ] [Y, ] 2 [X, ] X SS(= ) [ ] ? [OK] 2 375 = 10 10 11 2 ( )
8 11 8 SS 12 1 < >= 12 SS SS 11 < SS >= SS < 13 13 [ ] F p p 3 (1)
