Linear Regression
Michael Lee
October 9, 2015
The following labs are from James, Witten, Hastie, and Tibshirani’s An Introduction to Statistical Learning, pp. 109-115.
This short tutorial goes through univariate and multivariate linear regression, i.e., with one response variable and one or multiple predictor variables, respectively. It also includes an example of interaction terms, where predictor variables are multiplied and essentially play the role of another single variable.
The following libraries include the datasets to be used.
library(ISLR); library(MASS)Simple Linear Regression
The MASS library includes the Boston dataset on median house value for 506 neighborhoods around Boston, MA. Some details on the dataset follow:
data(Boston)
names(Boston)## [1] "crim" "zn" "indus" "chas" "nox" "rm" "age"
## [8] "dis" "rad" "tax" "ptratio" "black" "lstat" "medv"The dataset has 14 variables, including median house value (“medv”), but in this part, we’ll use only one predictor variable. First we try “lstat”, which is the percentage of households with low socioeconomic status. We want to see whether data on socioeconomic status is sufficiently predictive of median house values.
lm.fit <- lm(medv ~ lstat, data = Boston)
summary(lm.fit) ##
## Call:
## lm(formula = medv ~ lstat, data = Boston)
##
## Residuals:
## Min 1Q Median 3Q Max
## -15.168 -3.990 -1.318 2.034 24.500
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 34.55384 0.56263 61.41 <2e-16 ***
## lstat -0.95005 0.03873 -24.53 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 6.216 on 504 degrees of freedom
## Multiple R-squared: 0.5441, Adjusted R-squared: 0.5432
## F-statistic: 601.6 on 1 and 504 DF, p-value: < 2.2e-16To determine the coefficients of this simple model and obtain a confidene interval for the estimates, we run:
coef(lm.fit)## (Intercept) lstat
## 34.5538409 -0.9500494confint(lm.fit) ## 2.5 % 97.5 %
## (Intercept) 33.448457 35.6592247
## lstat -1.026148 -0.8739505Assuming we accept the predictive value of the model, we can use the predict() funtion to generate prediction and confidence intervals. The predictive interval incorporates both the irreducible and reducible error of the model, whereas the confidence interval minimizes the reducible error.
predict(lm.fit, data.frame(lstat=(c(5,10,15))),
interval = "prediction")## fit lwr upr
## 1 29.80359 17.565675 42.04151
## 2 25.05335 12.827626 37.27907
## 3 20.30310 8.077742 32.52846predict(lm.fit, data.frame(lstat=(c(5,10,15))),
interval = "confidence") ## fit lwr upr
## 1 29.80359 29.00741 30.59978
## 2 25.05335 24.47413 25.63256
## 3 20.30310 19.73159 20.87461Unsurprisingly, the prediction intervals (represented by the second and third columns) are much wider than their counterparts.
We plot “medv” against “lstat” with a scatterplot and the corresponding least squares regression line.
attach(Boston)
plot(lstat, medv, pch=20)
abline(lm.fit,lwd=3,col="red") 
There’s some indication that the relationship between lstat and medv isn’t linear. This will be discussed in the parts below.
Now we construct some diagnostic plots: the original plot.
par(mfrow=c(2,2))
plot(lm.fit)
We can get similar information by calling the plot() function with different arguments, as follows:
par(mfrow=c(1,1))
plot(predict(lm.fit), residuals(lm.fit))
plot(predict(lm.fit), rstudent(lm.fit))
plot(hatvalues(lm.fit))
which.max(hatvalues(lm.fit)) #Identifies which observation has the largest leverage statistic ## 375
## 375Multiple Linear Regression
We use the “lstat” and the “age” variables as predictors for the response variable “medv”. The second model uses all 13 variables.
lm.fit2 <- lm(medv ~ lstat + age, data=Boston)
summary(lm.fit2) ##
## Call:
## lm(formula = medv ~ lstat + age, data = Boston)
##
## Residuals:
## Min 1Q Median 3Q Max
## -15.981 -3.978 -1.283 1.968 23.158
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 33.22276 0.73085 45.458 < 2e-16 ***
## lstat -1.03207 0.04819 -21.416 < 2e-16 ***
## age 0.03454 0.01223 2.826 0.00491 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 6.173 on 503 degrees of freedom
## Multiple R-squared: 0.5513, Adjusted R-squared: 0.5495
## F-statistic: 309 on 2 and 503 DF, p-value: < 2.2e-16lm.fit13 <- lm(medv ~ ., data=Boston)
summary(lm.fit13)##
## Call:
## lm(formula = medv ~ ., data = Boston)
##
## Residuals:
## Min 1Q Median 3Q Max
## -15.595 -2.730 -0.518 1.777 26.199
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 3.646e+01 5.103e+00 7.144 3.28e-12 ***
## crim -1.080e-01 3.286e-02 -3.287 0.001087 **
## zn 4.642e-02 1.373e-02 3.382 0.000778 ***
## indus 2.056e-02 6.150e-02 0.334 0.738288
## chas 2.687e+00 8.616e-01 3.118 0.001925 **
## nox -1.777e+01 3.820e+00 -4.651 4.25e-06 ***
## rm 3.810e+00 4.179e-01 9.116 < 2e-16 ***
## age 6.922e-04 1.321e-02 0.052 0.958229
## dis -1.476e+00 1.995e-01 -7.398 6.01e-13 ***
## rad 3.060e-01 6.635e-02 4.613 5.07e-06 ***
## tax -1.233e-02 3.760e-03 -3.280 0.001112 **
## ptratio -9.527e-01 1.308e-01 -7.283 1.31e-12 ***
## black 9.312e-03 2.686e-03 3.467 0.000573 ***
## lstat -5.248e-01 5.072e-02 -10.347 < 2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 4.745 on 492 degrees of freedom
## Multiple R-squared: 0.7406, Adjusted R-squared: 0.7338
## F-statistic: 108.1 on 13 and 492 DF, p-value: < 2.2e-16Notice that the “age” variable in the lm.fit13 model has a high p-value. We can thus exclude it to see if we get better results.
lm.fit12 <- lm(medv ~ . -age, data=Boston)
summary(lm.fit12)##
## Call:
## lm(formula = medv ~ . - age, data = Boston)
##
## Residuals:
## Min 1Q Median 3Q Max
## -15.6054 -2.7313 -0.5188 1.7601 26.2243
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 36.436927 5.080119 7.172 2.72e-12 ***
## crim -0.108006 0.032832 -3.290 0.001075 **
## zn 0.046334 0.013613 3.404 0.000719 ***
## indus 0.020562 0.061433 0.335 0.737989
## chas 2.689026 0.859598 3.128 0.001863 **
## nox -17.713540 3.679308 -4.814 1.97e-06 ***
## rm 3.814394 0.408480 9.338 < 2e-16 ***
## dis -1.478612 0.190611 -7.757 5.03e-14 ***
## rad 0.305786 0.066089 4.627 4.75e-06 ***
## tax -0.012329 0.003755 -3.283 0.001099 **
## ptratio -0.952211 0.130294 -7.308 1.10e-12 ***
## black 0.009321 0.002678 3.481 0.000544 ***
## lstat -0.523852 0.047625 -10.999 < 2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 4.74 on 493 degrees of freedom
## Multiple R-squared: 0.7406, Adjusted R-squared: 0.7343
## F-statistic: 117.3 on 12 and 493 DF, p-value: < 2.2e-16Interaction Terms
We can create interaction terms, which essentially create a new predictor by multiplying existing predictors.
summary(lm(medv~lstat*age,data=Boston))##
## Call:
## lm(formula = medv ~ lstat * age, data = Boston)
##
## Residuals:
## Min 1Q Median 3Q Max
## -15.806 -4.045 -1.333 2.085 27.552
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 36.0885359 1.4698355 24.553 < 2e-16 ***
## lstat -1.3921168 0.1674555 -8.313 8.78e-16 ***
## age -0.0007209 0.0198792 -0.036 0.9711
## lstat:age 0.0041560 0.0018518 2.244 0.0252 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 6.149 on 502 degrees of freedom
## Multiple R-squared: 0.5557, Adjusted R-squared: 0.5531
## F-statistic: 209.3 on 3 and 502 DF, p-value: < 2.2e-16