1) Use the petrol consumption data set (http://myweb.fsu.edu/jelsner/temp/data/PetrolConsumption.txt) and build a regression tree to predict petrol consumption based on petrol tax, average income, amount of pavement and the proportion of the population with drivers licences. Plot the tree. Prune the tree leaving only three terminal nodes. Plot the final tree. (20)
setwd("C:/Spatial Statistics")
PetCon = read.table("http://myweb.fsu.edu/jelsner/temp/data/PetrolConsumption.txt", header = TRUE)
head(PetCon)
## Petrol.Tax Avg.Inc Pavement Prop.DL Petrol.Consumption
## 1 9.0 3571 1976 0.525 541
## 2 9.0 4092 1250 0.572 524
## 3 9.0 3865 1586 0.580 561
## 4 7.5 4870 2351 0.529 414
## 5 8.0 4399 431 0.544 410
## 6 10.0 5342 1333 0.571 457
suppressPackageStartupMessages(library(tree))
tr = tree(Petrol.Consumption ~ . , data = PetCon)
plot(tr)
text(tr)
tr1 = prune.tree(tr, best = 3)
plot(tr1)
text(tr1)
2) Use the data below to model the probability of O-ring damage as a logistic regression using launch temperature as the explanatory variable. Is the temperature a significant predictor of damage? Is it adequate? What are the odds of damage when launch temperature is 60F relative to the odds of damage when the temperature is 75F? Use the model to predict the probability of damage given a launch temperature of 55F. (30)
Temp = c(66, 70, 69, 68, 67, 72, 73, 70, 57, 63, 70, 78, 67, 53, 67, 75, 70, 81, 76, 79, 75, 76, 58)
Damage = c(0, 1, 0, 0, 0, 0, 0, 0, 1, 1, 1, 0, 0, 1, 0, 0, 0, 0, 0, 0, 1, 0, 1)
TempD = data.frame(Temp , Damage)
logrm = glm(Damage ~ Temp, data = TempD, family = binomial)
summary(logrm)
##
## Call:
## glm(formula = Damage ~ Temp, family = binomial, data = TempD)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -1.0611 -0.7613 -0.3783 0.4524 2.2175
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) 15.0429 7.3786 2.039 0.0415 *
## Temp -0.2322 0.1082 -2.145 0.0320 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 28.267 on 22 degrees of freedom
## Residual deviance: 20.315 on 21 degrees of freedom
## AIC: 24.315
##
## Number of Fisher Scoring iterations: 5
Given the p value of 0.03 there is moderate evidence that temperature is a major predictor of damage.
The change in deviance from the null to the full model is:
28.267 - 20.315 = 7.952 on 22 - 21 = 1 degree of freedom. Comparing this drop in deviance to a chi-squared value.
pchisq(7.952, 1, lower.tail = FALSE)
## [1] 0.004803426
The p-value for the model is 0.0048. Since it is less than 0.01, we can conclude that there is strong evidence that the model is significant and, hence, adequate in explaining damage.
You write the logistic regression model, where pi represents damage probability: \[ \hbox{logit}(\pi) = 15.043 - .0232 \times \hbox{Temp} \] In words: The logarithm of the odds of Damage equals 15.043 minus 0.232 times Temp.
So, the odds of damage when launch temperature is 60F relative to the odds of damage when the temperature is 75F is :
\[ \exp(-.0232 \times (60 - 75)) \] Or, exp(-0.232*(60 - 75)) = 32.460 ~ 32.5
As a result, the probabilities of damage at a launch temperature of 60F are approximately 32.5 times more than those at a launch temperature of 75F.
predict(logrm, data.frame(Temp = 55), type = "response")
## 1
## 0.9066965
Given a launch temperature of 55°F, we can forecast a damage probability of 90.7%based on the data and the model.
##3) Consider a set of medical records for 81 children undergoing a spinal operation. The data are in data frame called kyphosis (rpart package). The variables are: Kyphosis: A binary variable indicating the presence/absence of a post-operative spinal deformity called Kyphosis. Age: The age of the child in months. Number: The number of vertebrae involved in the spinal operation. Start: The beginning of the range of the vertebrae involved in the operation. (50)
suppressPackageStartupMessages(library(rpart))
head(kyphosis)
## Kyphosis Age Number Start
## 1 absent 71 3 5
## 2 absent 158 3 14
## 3 present 128 4 5
## 4 absent 2 5 1
## 5 absent 1 4 15
## 6 absent 1 2 16
names(kyphosis)
## [1] "Kyphosis" "Age" "Number" "Start"
mean(kyphosis$Age)/12
## [1] 6.971193
table(kyphosis$Kyphosis)
##
## absent present
## 64 17
sum(kyphosis$Kyphosis == "present")/length(kyphosis$Kyphosis)
## [1] 0.2098765
suppressPackageStartupMessages(library(ggplot2))
ggplot(kyphosis, aes(x = Kyphosis, y = Age)) + geom_boxplot() + ylab("Age (months")
logrmK = glm(Kyphosis ~ ., data =kyphosis, family = binomial)
summary(logrmK)
##
## Call:
## glm(formula = Kyphosis ~ ., family = binomial, data = kyphosis)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -2.3124 -0.5484 -0.3632 -0.1659 2.1613
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -2.036934 1.449575 -1.405 0.15996
## Age 0.010930 0.006446 1.696 0.08996 .
## Number 0.410601 0.224861 1.826 0.06785 .
## Start -0.206510 0.067699 -3.050 0.00229 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 83.234 on 80 degrees of freedom
## Residual deviance: 61.380 on 77 degrees of freedom
## AIC: 69.38
##
## Number of Fisher Scoring iterations: 5