Practice Assignment 7
Troy Phillips
10/15/2017
Diamonds are Expensive
Loading the libraries needed for this exercise, create a tibble out of diamonds using dplyr, and take a look at a summary to see what we have.
#Load all my stuff
#Rmd
#r message=FALSE, echo=FALSE, warning=FALSE
library(ggplot2)
library(dplyr)
library(magrittr)
library(ggthemes)
#see what we have
#Create a tibble using dplyr
tbl_diamond <- tbl_df(diamonds)
summary(tbl_diamond) carat cut color clarity
Min. :0.2000 Fair : 1610 D: 6775 SI1 :13065
1st Qu.:0.4000 Good : 4906 E: 9797 VS2 :12258
Median :0.7000 Very Good:12082 F: 9542 SI2 : 9194
Mean :0.7979 Premium :13791 G:11292 VS1 : 8171
3rd Qu.:1.0400 Ideal :21551 H: 8304 VVS2 : 5066
Max. :5.0100 I: 5422 VVS1 : 3655
J: 2808 (Other): 2531
depth table price x
Min. :43.00 Min. :43.00 Min. : 326 Min. : 0.000
1st Qu.:61.00 1st Qu.:56.00 1st Qu.: 950 1st Qu.: 4.710
Median :61.80 Median :57.00 Median : 2401 Median : 5.700
Mean :61.75 Mean :57.46 Mean : 3933 Mean : 5.731
3rd Qu.:62.50 3rd Qu.:59.00 3rd Qu.: 5324 3rd Qu.: 6.540
Max. :79.00 Max. :95.00 Max. :18823 Max. :10.740
y z
Min. : 0.000 Min. : 0.000
1st Qu.: 4.720 1st Qu.: 2.910
Median : 5.710 Median : 3.530
Mean : 5.735 Mean : 3.539
3rd Qu.: 6.540 3rd Qu.: 4.040
Max. :58.900 Max. :31.800
#With this much data, we need to make sure R sees clarity as a factor for this graph
Clarity <- as.factor(diamonds$clarity)Step 1: Use a dot-plot to compare price, carats, and clarity
This shows that the more the clarity increases and carats increases, the price increases…fast. For reference, the clarity chart:
a <- tbl_diamond %>%
ggplot(aes(x=carat, y=price, color= Clarity)) +
geom_point()+
xlab("Carats")+
ylab("Price")+
guides(fill = guide_legend("Clarity"))+
ggtitle("Comparing Diamond Cut, Price and Clarity")
#be patient, there is alot of data to plot. This takes a second or two.
a
Step 2: Use a dot-plot to compare cut, price, and carat
This shows that while some cuts may have higher prices, the ‘ideal’ cut is all over the place.
#make cut a factor for mapping a third variable
fact_cut <- as.factor(diamonds$cut)
#Compare cut and carat with price
b <- tbl_diamond %>%
ggplot(aes(x=carat, y=price, color=fact_cut)) +
geom_point() +
xlab("Carats")+
ylab("Price")+
ggtitle("Comparing Diamond Carat and Price by Cut")+
scale_colour_discrete (name="Cut")
b
Step 3: Ideal Cut Distribution
We pull out the ideal cut to see the distribution. It still appears there is nothing special about the ideal cut.
#What is 'ideal'?
#First, let's filter out 'ideal' from the data
tbl_diamond_no_ideal <- tbl_diamond %>%
filter(cut != "Ideal")
#and make a new factor
fact_cut_noideal <- as.factor((tbl_diamond_no_ideal$cut))
#Next, filter for only showing 'ideal'
tbl_diamond_ideal <- tbl_diamond %>%
filter(cut == "Ideal")
#Look at the Ideal cut compared to carat and price
c_ideal <- tbl_diamond_ideal %>%
ggplot(aes(x=carat, y=price)) +
geom_point()+
xlab("Carats")+
ylab("Price")+
guides(fill = guide_legend("Clarity"))+
ggtitle("Comparing Diamond Carat and Price for Ideal cuts")+
theme_stata()
c_ideal
Step 4: Compare Price and Cut
This boxplot shows the quartiles and mean for price of the different cuts. It doesn’t look like the ‘better’ cuts have a significantly higher average price, as I would have expected.
# Make a boxplot comparing price and cut
d_box1 <- tbl_diamond %>%
ggplot(aes(x = fact_cut, y = price))+
geom_boxplot()+
xlab("Cut")+
ylab("Price")+
ggtitle("Boxplot to see Cut and Price")+
theme_solarized_2()
d_box1
Step 5: Compare Price and Clarity
This boxplot shows the quartiles and mean for price of the different levels of clarity. Interstingly, the mean prices for the better clarity is not lower. In fact, it looks like it goes down in some instances.
d_box2 <- tbl_diamond %>%
ggplot(aes(x=clarity, y=price)) +
geom_boxplot() +
xlab("Clarity")+
ylab("Price")+
ggtitle("Comparing Diamond Clarity and Price")+
theme_solarized_2()
d_box2
Step 6: Compare Carat and Clarity
Let’s compare the carat and clarity to see if there is an indication as to what is going on. It looks like as clarity goes up, the average carat goes down!
d_box3 <- tbl_diamond %>%
ggplot(aes(x=clarity, y=carat)) +
geom_boxplot()+
xlab("Clarity")+
ylab("Carat")+
ggtitle("Comparing Diamond Carat and Clarity")+
theme_solarized_2()
d_box3
Step 7: Line Graph of Mean Price by Cut
Do some analysis on the data and create a line graph the mean price by cut. We normalize the carat by dividing the price by carat to get Price Per Carat. We see that the price per carat does increase as cut gets “better”, until we get to the “ideal” cut…which seems to be ideal if you are trying to save money!
##Let's see what the average carat by clarity is. This may explain the price.
gb_clarity <- group_by(tbl_diamond, clarity)
summarise(gb_clarity, mean(carat, na.rm=TRUE))# A tibble: 8 x 2
clarity `mean(carat, na.rm = TRUE)`
<ord> <dbl>
1 I1 1.2838462
2 SI2 1.0776485
3 SI1 0.8504822
4 VS2 0.7639346
5 VS1 0.7271582
6 VVS2 0.5962021
7 VVS1 0.5033215
8 IF 0.5051229#Normalize the price to compare different sized (carat) diamonds
tbl_mutate_caratprice <- tbl_diamond %>%
mutate(mut_caratprice = price/carat)
##For all cuts
gb_clarity_ppc <- group_by(tbl_mutate_caratprice, clarity)
sum_clarity_ppc <- summarise(gb_clarity_ppc, mPrice = mean(price), mPPC = mean(mut_caratprice, na.rm=TRUE))
gb_cut_ppc <- group_by(tbl_mutate_caratprice, cut)
sum_cut_PPC <- summarise(gb_cut_ppc, mPrice = mean(price), mPPC = mean(mut_caratprice, na.rm = TRUE))
summary(sum_cut_PPC) cut mPrice mPPC
Fair :1 Min. :3458 Min. :3767
Good :1 1st Qu.:3929 1st Qu.:3860
Very Good:1 Median :3982 Median :3920
Premium :1 Mean :4062 Mean :3957
Ideal :1 3rd Qu.:4359 3rd Qu.:4014
Max. :4584 Max. :4223 line_cut_PPC <- sum_cut_PPC %>%
ggplot(aes(x = cut, y = mPPC, group = 1 ))+
geom_point(size = 3) +
geom_line(size = 2)+
xlab("Diamond Cut")+
ylab("Average Price per Carat")+
ggtitle("Average Price Per Carat by Cut: is Ideal, Ideal?")+
theme(legend.position = "bottom")+
theme_solarized()
line_cut_PPC