Questions #3, 5, 6, 9
Q3. We now review k-fold cross-validation.
K-fold cross-validation involves randomly dividing the set of observations into k groups (folds), of approximately equal size. The first fold is treated as a validation set, and the remainder is treated as a training set. The k-fold CV estimate is estimated by averaging the k resulting MSE estimates.
The validation set approach is simple as you randomly divide the available set of observations into two parts: training set and validation set (or hold-out set). This approach can be highly variable depending on which observations are included in the training set and which are included in the validation set; may tend to overestimate the test error rate for the model fit on the entire data set.
LOOCV is similar to validation set approach as it attempts to address the method’s drawbacks/disadvantages by using a single observation for the validation set and the remaining observations for the training set. It has far less bias and does not overestimate the test error rate as much as the validation set does, and yields the same result when performing LOOCV multiple times. However, it is a poor estimate because it is highly variable since it is based upon a single observation and can potentially be expensive to implement because the model has to be fit n times.
Q5. In Chapter 4, we used logistic regression to predict the probability of default using income and balance on the Default data set. We will now estimate the test error of this logistic regression model using the validation set approach. Do not forget to set a random seed before beginning your analysis.
library(ISLR)
attach(Default)
set.seed(1)
glm.fit = glm(default ~ income + balance,
data = Default,
family = "binomial")
summary(glm.fit)
##
## Call:
## glm(formula = default ~ income + balance, family = "binomial",
## data = Default)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -2.4725 -0.1444 -0.0574 -0.0211 3.7245
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.154e+01 4.348e-01 -26.545 < 2e-16 ***
## income 2.081e-05 4.985e-06 4.174 2.99e-05 ***
## balance 5.647e-03 2.274e-04 24.836 < 2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 2920.6 on 9999 degrees of freedom
## Residual deviance: 1579.0 on 9997 degrees of freedom
## AIC: 1585
##
## Number of Fisher Scoring iterations: 8
train = sample(dim(Default)[1], dim(Default)[1] / 2)
glm.fit = glm(default ~ income + balance,
data = Default,
family = "binomial",
subset = train)
summary(glm.fit)
##
## Call:
## glm(formula = default ~ income + balance, family = "binomial",
## data = Default, subset = train)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -2.5830 -0.1428 -0.0573 -0.0213 3.3395
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.194e+01 6.178e-01 -19.333 < 2e-16 ***
## income 3.262e-05 7.024e-06 4.644 3.41e-06 ***
## balance 5.689e-03 3.158e-04 18.014 < 2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 1523.8 on 4999 degrees of freedom
## Residual deviance: 803.3 on 4997 degrees of freedom
## AIC: 809.3
##
## Number of Fisher Scoring iterations: 8
glm.probs = predict(glm.fit,
newdata = Default[-train, ],
type = "response")
glm.pred = rep("No", length(glm.probs))
glm.pred[glm.probs > 0.5] = "Yes"
mean(glm.pred != Default[-train, ]$default)
## [1] 0.0254
The three different splits each produce a different resulting error rate which shows that the rate varies by which observations are in the training/validation sets.
#1
train = sample(dim(Default)[1], dim(Default)[1] / 2)
glm.fit = glm(default ~ income + balance,
data = Default,
family = "binomial",
subset = train)
glm.probs = predict(glm.fit,
newdata = Default[-train, ],
type = "response")
glm.pred = rep("No", length(glm.probs))
glm.pred[glm.probs > 0.5] = "Yes"
mean(glm.pred != Default[-train, ]$default)
## [1] 0.0274
#2
train = sample(dim(Default)[1], dim(Default)[1] / 2)
fit.glm = glm(default ~ income + balance,
data = Default,
family = "binomial",
subset = train)
glm.probs = predict(fit.glm,
newdata = Default[-train, ],
type = "response")
glm.pred = rep("No", length(glm.probs))
glm.pred[glm.probs > 0.5] = "Yes"
mean(glm.pred != Default[-train, ]$default)
## [1] 0.0244
#3
train = sample(dim(Default)[1], dim(Default)[1] / 2)
fit.glm = glm(default ~ income + balance,
data = Default,
family = "binomial",
subset = train)
glm.probs = predict(fit.glm,
newdata = Default[-train, ],
type = "response")
glm.pred = rep("No", length(glm.probs))
glm.pred[glm.probs > 0.5] = "Yes"
mean(glm.pred != Default[-train, ]$default)
## [1] 0.0244
Including the dummy variable did not result in a reduction in the test rate.
train = sample(dim(Default)[1], dim(Default)[1] / 2)
glm.fit = glm(default ~ income + balance + student,
data = Default,
family = "binomial",
subset = train)
glm.probs = predict(fit.glm,
newdata = Default[-train, ],
type = "response")
glm.pred = rep("No", length(glm.probs))
glm.pred[glm.probs > 0.5] = "Yes"
mean(glm.pred != Default[-train, ]$default)
## [1] 0.0266
Q6. We continue to consider the use of a logistic regression model to predict the probability of default using income and balance on the Default data set. In particular, we will now compute estimates for the standard errors of the income and balance logistic regression coefficients in two different ways: (1) using the bootstrap, and (2) using the standard formula for computing the standard errors in the glm() function. Do not forget to set a random seed before beginning your analysis.
set.seed(1)
train = sample(dim(Default)[1], dim(Default)[1] / 2)
glm.fit <- glm(default ~ income + balance,
data = Default,
family = "binomial")
summary(glm.fit)
##
## Call:
## glm(formula = default ~ income + balance, family = "binomial",
## data = Default)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -2.4725 -0.1444 -0.0574 -0.0211 3.7245
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.154e+01 4.348e-01 -26.545 < 2e-16 ***
## income 2.081e-05 4.985e-06 4.174 2.99e-05 ***
## balance 5.647e-03 2.274e-04 24.836 < 2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 2920.6 on 9999 degrees of freedom
## Residual deviance: 1579.0 on 9997 degrees of freedom
## AIC: 1585
##
## Number of Fisher Scoring iterations: 8
boot.fn = function(data, index) {
fit = glm(default ~ income + balance,
data = data,
family = "binomial",
subset = index)
return (coef(fit))
}
The bootstrap estimates of the standard errors of the coefficients are β1 and β2 are respectively 4.858 x 10^(-6) and 2.300 x 10^(-4).
library(boot)
boot(Default, boot.fn, 1000)
##
## ORDINARY NONPARAMETRIC BOOTSTRAP
##
##
## Call:
## boot(data = Default, statistic = boot.fn, R = 1000)
##
##
## Bootstrap Statistics :
## original bias std. error
## t1* -1.154047e+01 -3.912114e-02 4.347403e-01
## t2* 2.080898e-05 1.585717e-07 4.858722e-06
## t3* 5.647103e-03 1.856917e-05 2.300758e-04
The estimated standard errors obtained are very similar/close.
detach(Default)
Q9. We will now consider the Boston housing data set, from the ISLR2 library.
library(ISLR2)
## Warning: package 'ISLR2' was built under R version 4.1.3
##
## Attaching package: 'ISLR2'
## The following objects are masked from 'package:ISLR':
##
## Auto, Credit
attach(Boston)
mu.hat = mean(medv)
mu.hat
## [1] 22.53281
se.hat = sd(medv) / sqrt(dim(Boston)[1])
se.hat
## [1] 0.4088611
The bootstrap estimated standard error from mu.hat is 0.4106622, which is similar to the estimate found from se.hat of 0.4088611.
set.seed(1)
boot.fn = function(data, index) {
mu = mean(data[index])
return (mu)
}
boot(medv, boot.fn, 1000)
##
## ORDINARY NONPARAMETRIC BOOTSTRAP
##
##
## Call:
## boot(data = medv, statistic = boot.fn, R = 1000)
##
##
## Bootstrap Statistics :
## original bias std. error
## t1* 22.53281 0.007650791 0.4106622
The bootstrap confidence interval is very close to the one provided by the t.test() function.
t.test(medv)
##
## One Sample t-test
##
## data: medv
## t = 55.111, df = 505, p-value < 2.2e-16
## alternative hypothesis: true mean is not equal to 0
## 95 percent confidence interval:
## 21.72953 23.33608
## sample estimates:
## mean of x
## 22.53281
CI.mu.hat = c(22.53 - 2 * 0.4106622, 22.53 + 2 * 0.4106622)
CI.mu.hat
## [1] 21.70868 23.35132
med.hat = median(medv)
med.hat
## [1] 21.2
Estimated median value of 21.2 is equal to the value obtained in (e); with a standard error of 0.3770241 which is relatively small.
boot.fn = function(data, index) {
mu = median(data[index])
return (mu)
}
boot(medv, boot.fn, 1000)
##
## ORDINARY NONPARAMETRIC BOOTSTRAP
##
##
## Call:
## boot(data = medv, statistic = boot.fn, R = 1000)
##
##
## Bootstrap Statistics :
## original bias std. error
## t1* 21.2 -0.0386 0.3770241
percent.hat = quantile(medv, c(0.1))
percent.hat
## 10%
## 12.75
Estimated 12.75 which is equal to the value obtained in (g); with a standard error of 0.4925766 which is relatively small.
boot.fn = function(data, index) {
mu = quantile(data[index], c(0.1))
return (mu)
}
boot(medv, boot.fn, 1000)
##
## ORDINARY NONPARAMETRIC BOOTSTRAP
##
##
## Call:
## boot(data = medv, statistic = boot.fn, R = 1000)
##
##
## Bootstrap Statistics :
## original bias std. error
## t1* 12.75 0.0186 0.4925766
detach(Boston)