API tools faq
paste
Guest User

Untitled

a guest
Nov 21st, 2017
70
0
Never
Add comment
Not a member of Pastebin yet? Sign Up, it unlocks many cool features!
text 6.26 KB | None | 0 0
raw download clone embed print report
  1. ctree <- ctree(quotation ~ minute + temp, data = visitquot)
  2. print(ctree)
  3.  
  4. Fitted party:
  5. [1] root
  6. | [2] minute <= 600
  7. | | [3] minute <= 227
  8. | | | [4] temp <= -0.4259
  9. | | | | [5] temp <= -2.3174: 0.015 (n = 6254, err = 89.7)
  10. | | | | [6] temp > -2.3174
  11. | | | | | [7] minute <= 68: 0.028 (n = 4562, err = 126.3)
  12. | | | | | [8] minute > 68: 0.046 (n = 7100, err = 312.8)
  13. | | | [9] temp > -0.4259
  14. | | | | [10] temp <= 6.0726: 0.015 (n = 56413, err = 860.5)
  15. | | | | [11] temp > 6.0726: 0.019 (n = 39779, err = 758.9)
  16. | | [12] minute > 227
  17. | | | [13] minute <= 501
  18. | | | | [14] minute <= 291: 0.013 (n = 30671, err = 388.0)
  19. | | | | [15] minute > 291: 0.009 (n = 559646, err = 5009.3)
  20. | | | [16] minute > 501
  21. | | | | [17] temp <= 5.2105
  22. | | | | | [18] temp <= -1.8393: 0.009 (n = 66326, err = 617.1)
  23. | | | | | [19] temp > -1.8393: 0.012 (n = 355986, err = 4289.0)
  24. | | | | [20] temp > 5.2105
  25. | | | | | [21] temp <= 13.6927: 0.014 (n = 287909, err = 3900.7)
  26. | | | | | [22] temp > 13.6927
  27. | | | | | | [23] temp <= 14: 0.035 (n = 2769, err = 92.7)
  28. | | | | | | [24] temp > 14: 0.007 (n = 2161, err = 15.9)
  29. | [25] minute > 600
  30. | | [26] temp <= 1.6418
  31. | | | [27] temp <= -2.3366: 0.012 (n = 110810, err = 1268.1)
  32. | | | [28] temp > -2.3366: 0.014 (n = 584457, err = 7973.2)
  33. | | [29] temp > 1.6418: 0.016 (n = 3753208, err = 57864.3)
  34.  
  35. plot(ctree, type = "simple")
  36.  
  37. library(partykit)
  38. airq <- subset(airquality, !is.na(Ozone))
  39. airct <- ctree(Ozone ~ ., data = airq)
  40. print(airct)
  41. # Model formula:
  42. # Ozone ~ Solar.R + Wind + Temp + Month + Day
  43. #
  44. # Fitted party:
  45. # [1] root
  46. # | [2] Temp <= 82
  47. # | | [3] Wind <= 6.9: 55.600 (n = 10, err = 21946.4)
  48. # | | [4] Wind > 6.9
  49. # | | | [5] Temp <= 77: 18.479 (n = 48, err = 3956.0)
  50. # | | | [6] Temp > 77: 31.143 (n = 21, err = 4620.6)
  51. # | [7] Temp > 82
  52. # | | [8] Wind <= 10.3: 81.633 (n = 30, err = 15119.0)
  53. # | | [9] Wind > 10.3: 48.714 (n = 7, err = 1183.4)
  54. #
  55. # Number of inner nodes: 4
  56. # Number of terminal nodes: 5
  57.  
  58. detach("package:partykit", unload=TRUE)
  59. library(party)
  60. airct <- party::ctree(Ozone ~ ., data = airq)
  61.  
  62. t(sapply(unique(where(airct)), function(x) {
  63. n <- nodes(airct, x)[[1]]
  64. Ozone <- airq[as.logical(n$weights), "Ozone"]
  65. cbind.data.frame("Node" = as.integer(x),
  66. "n" = length(Ozone),
  67. "Avg."= mean(Ozone),
  68. "Variance"= var(Ozone),
  69. "SSE" = sum((Ozone - mean(Ozone))^2))
  70. }))
  71.  
  72. # Node n Avg. Variance SSE
  73. # [1,] 5 48 18.47917 84.16977 3955.979
  74. # [2,] 3 10 55.6 2438.489 21946.4
  75. # [3,] 6 21 31.14286 231.0286 4620.571
  76. # [4,] 9 7 48.71429 197.2381 1183.429
  77. # [5,] 8 30 81.63333 521.3437 15118.97
  78.  
  79. terNodes <- unique(where(airct))
  80. setdiff(1:max(terNodes), terNodes)
  81.  
  82. sapply(setdiff(1:max(terNodes), terNodes), function(x) {
  83. n <- nodes(airct, x)[[1]]
  84. pvalue <- 1 - nodes(airct, x)[[1]]$criterion$maxcriterion
  85. plab <- ifelse(pvalue < 10^(-3),
  86. paste("p <", 10^(-3)),
  87. paste("p =", round(pvalue, digits = 3)))
  88. c("Node" = x, "P-value" = plab)
  89. })
  90.  
  91. # [,1] [,2] [,3] [,4]
  92. # Node "1" "2" "4" "7"
  93. # P-value "p < 0.001" "p = 0.002" "p = 0.003" "p = 0.003"
  94.  
  95. library("partykit")
  96. data("PimaIndiansDiabetes", package = "mlbench")
  97. ct <- ctree(diabetes ~ ., data = PimaIndiansDiabetes)
  98. plot(as.simpleparty(ct))
  99.  
  100. nodeapply(ct, ids = 1, function(n) info_node(n)$criterion)
  101. ## $`1`
  102. ## pregnant glucose pressure triceps insulin
  103. ## statistic 3.631413e+00 5.117841e+00 1.1778530 1.455334 2.570457503
  104. ## p.value -6.380290e-09 -2.710725e-37 -0.5937987 -0.313498 -0.002398554
  105. ## mass pedigree age
  106. ## statistic 4.185236e+00 3.143294e+00 3.774507e+00
  107. ## p.value -4.181295e-15 -1.180135e-05 -3.262459e-10
  108.  
  109. library("strucchange")
  110. sctest(ct, node = 1)
  111. ## pregnant glucose pressure triceps insulin mass
  112. ## statistic 3.776615e+01 166.9745 3.2473947 4.2859164 13.07180346 6.570902e+01
  113. ## p.value 6.380290e-09 0.0000 0.4477744 0.2691142 0.00239568 4.218847e-15
  114. ## pedigree age
  115. ## statistic 2.318009e+01 4.357601e+01
  116. ## p.value 1.180128e-05 3.262459e-10
  117.  
  118. nodeapply(ct, ids = nodeids(ct), function(n) info_node(n)$p.value)
  119. ## $`1`
  120. ## glucose
  121. ## 0
  122. ##
  123. ## $`2`
  124. ## age
  125. ## 6.048661e-07
  126. ##
  127. ## $`3`
  128. ## mass
  129. ## 0.001169778
  130. ##
  131. ## ...
  132.  
  133. tab <- tapply(PimaIndiansDiabetes$diabetes, predict(ct, type = "node"),
  134. function(y) c("n" = length(y), 100 * prop.table(table(y))))
  135. do.call("rbind", tab)
  136. ## n neg pos
  137. ## 5 144 99.30556 0.6944444
  138. ## 6 7 85.71429 14.2857143
  139. ## 7 120 82.50000 17.5000000
  140. ## 8 214 66.82243 33.1775701
  141. ## 11 53 79.24528 20.7547170
  142. ## 12 108 39.81481 60.1851852
  143. ## 13 122 19.67213 80.3278689
  144.  
  145. library(partykit)
  146.  
  147. # data
  148. airquality = data.frame(airquality)
  149.  
  150. # create a numeric binary variable as dependent variable
  151. airquality$OzoneClass = 0
  152. airquality$OzoneClass[airquality$Ozone>=34] =1
  153.  
  154. # regression tree with scale dependent variable
  155. airq <- subset(airquality, !is.na(Ozone))
  156. airct <- ctree(OzoneClass ~ Temp, data = airq)
  157. print(airct)
  158.  
  159. # Model formula:
  160. # OzoneClass ~ Temp
  161. #
  162. # Fitted party:
  163. # [1] root
  164. # | [2] Temp <= 82
  165. # | | [3] Temp <= 77: 0.096 (n = 52, err = 4.5)
  166. # | | [4] Temp > 77: 0.519 (n = 27, err = 6.7)
  167. # | [5] Temp > 82: 0.973 (n = 37, err = 1.0)
  168. #
  169. # Number of inner nodes: 2
  170. # Number of terminal nodes: 3
  171.  
  172.  
  173.  
  174.  
  175. # create a categorical binary variable as dependent variable
  176. airquality$OzoneClass = 0
  177. airquality$OzoneClass[airquality$Ozone>=34] =1
  178. airquality$OzoneClass = as.factor(airquality$OzoneClass)
  179.  
  180. # classification tree
  181. airq <- subset(airquality, !is.na(Ozone))
  182. airct <- ctree(OzoneClass ~ Temp, data = airq)
  183. print(airct)
  184.  
  185. # Model formula:
  186. # OzoneClass ~ Temp
  187. #
  188. # Fitted party:
  189. # [1] root
  190. # | [2] Temp <= 82
  191. # | | [3] Temp <= 77: 0 (n = 52, err = 9.6%)
  192. # | | [4] Temp > 77: 1 (n = 27, err = 48.1%)
  193. # | [5] Temp > 82: 1 (n = 37, err = 2.7%)
  194. #
  195. # Number of inner nodes: 2
  196. # Number of terminal nodes: 3
Add Comment
Please, Sign In to add comment
Public Pastes
We use cookies for various purposes including analytics. By continuing to use Pastebin, you agree to our use of cookies as described in the Cookies Policy.  OK, I Understand
Not a member of Pastebin yet?
Sign Up, it unlocks many cool features!