Can you price a used Honda Accord just by reading its odometer? In this analysis, we explore the relationship between a used Honda Accord’s mileage (in thousands of miles) and its price (in thousands of dollars).
library(Stat2Data)
library(dplyr)
library(ggplot2)
data(AccordPrice)
my_data <- AccordPrice
We fit a linear model where Mileage is the explanatory
variable (\(x\)) and Price
is the response variable (\(y\)).
ggplot(my_data, aes(x = Mileage, y = Price)) +
geom_point() +
labs(
title = "Price of Used Honda Accords vs. Mileage",
x = "Mileage (thousands of miles)",
y = "Price (thousands of dollars)"
)
cor(my_data$Mileage, my_data$Price)
## [1] -0.8489441
The strong negative correlation (\(r = -0.8489\)) indicates that as mileage increases, the price of a used Honda Accord tends to decrease.
model <- lm(Price ~ Mileage, data = my_data)
coef(model)
## (Intercept) Mileage
## 20.8096129 -0.1198119
Below is the scatterplot with our regression line added:
ggplot(my_data, aes(x = Mileage, y = Price)) +
geom_point() +
geom_abline(intercept = 20.8096129, slope = -0.1198119, color = "steelblue") +
labs(
title = "Price of Used Honda Accords vs. Mileage with Fitted Line",
x = "Mileage (thousands of miles)",
y = "Price (thousands of dollars)"
)
We examined the first observation in the dataset:
my_data %>%
select(Mileage, Price) %>%
head(1)
## Mileage Price
## 1 74.9 12
For the first row (Mileage = 74.9 thousand miles, Price = 12 thousand dollars), we computed the residual by hand:
12 - (-0.1198119 * 74.9 + 20.8096129)
## [1] 0.1642984
Interpretation: The residual is \(+0.1643\) (thousands of dollars), meaning this car’s actual price was $164.30 higher than predicted by the regression line.
We added predicted values and residuals to our dataset and generated a residual plot to check model assumptions:
my_data <- my_data %>%
mutate(
predicted = fitted(model),
residual = residuals(model)
)
my_data %>%
select(Mileage, Price, predicted, residual) %>%
head()
## Mileage Price predicted residual
## 1 74.9 12.0 11.83570 0.1643021
## 2 53.0 17.9 14.45958 3.4404204
## 3 79.1 15.7 11.33249 4.3675123
## 4 50.1 12.5 14.80703 -2.3070342
## 5 62.0 9.5 13.38127 -3.8812720
## 6 4.8 21.5 20.23452 1.2654845
ggplot(my_data, aes(x = Mileage, y = residual)) +
geom_point(alpha = 0.6) +
geom_hline(yintercept = 0, color = "steelblue") +
labs(
title = "Residual Plot for Honda Accord Price Model",
x = "Mileage (thousands of miles)",
y = "Residual (thousands of dollars)"
)
Interpretation: The residual plot is “boring”—a random scatter of points around zero with no clear curve or fanning pattern. This confirms that a straight-line model is appropriate for this dataset.
Finally, we evaluated the summary of the model:
summary(model)
##
## Call:
## lm(formula = Price ~ Mileage, data = my_data)
##
## Residuals:
## Min 1Q Median 3Q Max
## -6.5984 -1.8169 -0.4148 1.4502 6.5655
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 20.8096 0.9529 21.84 < 2e-16 ***
## Mileage -0.1198 0.0141 -8.50 3.06e-09 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 3.085 on 28 degrees of freedom
## Multiple R-squared: 0.7207, Adjusted R-squared: 0.7107
## F-statistic: 72.25 on 1 and 28 DF, p-value: 3.055e-09
Interpretation: 1. \(R^2\): \(72.07\%\) of the variation in Honda Accord prices is explained by the linear relationship with mileage.