8.1 Correlation plots
David Welch
10/8/2019
In this exercise you will learn to visualize the pairwise relationships between a set of quantitative variables. To this end, you will make your own note of 8.1 Correlation plots from Data Visualization with R.
# import data
data(SaratogaHouses, package="mosaicData")
# select numeric variables
df <- dplyr::select_if(SaratogaHouses, is.numeric)
# calulate the correlations
r <- cor(df, use="complete.obs")
round(r,2)library(ggplot2)
library(ggcorrplot)
# visualize the correlations
ggcorrplot(r,
hc.order = TRUE,
type = "lower",
lab = TRUE)
Q1 What factors have positve correlation with home price?
The more fireplaces a house gets the omre expensive it is.
Q2 What factors have strong positve correlation with home price?
Living area has a strong correlation with home price. The bigger the rooms the more expensive the house is.
Q3 What factors have negative correlation with home price?
Age has has a negative correlation with home price.
Q4 What factors have strong negative correlation with home price?
There are no factors that have a strong negative correlation with home price.
Q5 What set of two variables has the highest positive Pearson Product-Moment correlation coefficient? What set of two variables has the greatest negative Pearson Product-Moment correlation coefficient?
Living area and rooms has the highest positive correlation corefficient. Bathrooms and age has the higihest negative correlation coefficient.
Q6 What set of two variables has the Pearson Product-Moment correlation coefficent that is closest to zero? Would you be sure that the two variables are not related at all? What would you do to check?
ggplot(SaratogaHouses,
aes(x = landValue,
y = age)) +
geom_point()
Q7 Plot correlation for CPS85 in the same way as above. Repeat Q1-Q6.
Hint: The CPS85 data set is from the mosaicData package. Explain wage instead of home price.
# import data
data(CPS85, package="mosaicData")
# select numeric variables
df <- dplyr::select_if(CPS85, is.numeric)
# calulate the correlations
r <- cor(df, use="complete.obs")
round(r,2)library(ggplot2)
library(ggcorrplot)
# visualize the correlations
ggcorrplot(r,
hc.order = TRUE,
type = "lower",
lab = TRUE)
Education, age, exper all has a positive correlation with wage.
No factors have a strong positive correlation with the wage.
No factors have a negative correlation with the wage.
No facotrs have a strong negative correlation with the wage.
Exper and age have the highest positve correlation coefficient.
Exper and educaton have the greatest negative correlation coefficient.
ggplot(CPS85,
aes(x = age,
y = wage)) +
geom_point()
Q8 Hide the messages, the code and its results on the webpage.
Hint: Use message, echo and results in the chunk options. Refer to the RMarkdown Reference Guide.