In the MinnLand dataset in the alr4 package, fit two possible models with log(acrePrice) or sqrt(acrePrice) as the response (y) variable and your choice of independent (x) variables. Be sure to...

Assignment 112062
Assignment 112062
In the MinnLand dataset in the alr4 package, fit two possible models with log(acrePrice) or sqrt(acrePrice) as the response variable and your choice of independent variables. We will be using log(acrePrice) in this exercise. Note that the interpretations won’t change even if we used sqrt(acrePrice).
Let us start the exercise by importing the required packages, i.e. alr4 and caret. The package caret is used for cross validation.
rm(list=ls(all=TRUE))
library(alr4)
## Loading required package: car
## Loading required package: carData
## Loading required package: effects
## lattice theme set by effectsTheme()
## See ?effectsTheme for details.
library(caret)
## Loading required package: ggplot2
## Loading required package: lattice
The first regression model we fit is between log(acrePrice) and year. It is well known that the land prices increases with time. This model can help us to confirm that notion.
## Regression between log(acrePrice) and year
plot(as.factor(MinnLand$year), MinnLand$acrePrice, xlab = "Year",
ylab = "Sale Price in Dollars per Acre",
main = "year vs. acrePrice")
lm(log(MinnLand$acrePrice) ~ as.factor(MinnLand$year))
##
## Call:
## lm(formula = log(MinnLand$acrePrice) ~ as.factor(MinnLand$year))
##
## Coefficients:
## (Intercept) as.factor(MinnLand$year)2003
## 7.27175 -0.00155
## as.factor(MinnLand$year)2004 as.factor(MinnLand$year)2005
## 0.14794 0.36026
## as.factor(MinnLand$year)2006 as.factor(MinnLand$year)2007
## 0.39392 0.47682
## as.factor(MinnLand$year)2008 as.factor(MinnLand$year)2009
## 0.68364 ...