ITEC 621 Exercise 1 - R Refresher

J. Alberto Espinosa

2025-01-04

This Quarto file contains ITEC 621 Exercise 1.

Table of contents
General Instructions........................................................................................................................................ 1
Quarto Overview (please read carefully)................................................................................................ 1
1. Basic R Concepts .......................................................................................................................................... 2
2. Data Manipulation ....................................................................................................................................... 3
3. Basic Descriptive Analytics ...................................................................................................................... 4
4. Basic Predictive Analytics ........................................................................................................................ 7

General Instructions
Download the Quarto template for this exercise Ex1_R_YourLastName.Qmd and save it
with your own last name exactly. Then open it in R Studio and complete all the exercises
and answer the questions below in the template. Run the code to ensure everything is
working fine. When done, knit your R Markdown file into a Word document and submit it.
No need to submit the .Qmd, file just the Word or PDF knitted file. If for some reason you
can’t knit a Word or PDF file, you can knit to an HTML file and then save it as a PDF. Some
LMS systems don’t accept HTML submissions and your HTML file may not display well
without its companion files folder.
This exercise is somewhat similar to HW0 in KSB-999, which you were required to
complete before starting this course. So, if you already did that, this should be an easy
exercise and a good warm up refresher. If you didn’t do it, this is you opportunity to catch
up. This course moves fast and it assumes that you have some familiarity with R.

Quarto Overview (please read carefully)

See full instructions on how to use Quarto in ITEC_Quarto.Qmd.
When you create a Quarto file, it will look like text commingled with R code. You can edit
your Quarto file in either Source or Visual mode, but clicking on the corresponding top left
button. Visual is OK for demos, but I highly encourage you to write your code in Source
view. You will also see a button option named Render in your tool bar (it will only show if
your file has the .Qmd extension. Once you are done with all the coding, click on the Render
button and Quarto will knit your document in the format specified in the YAML, with all
your marked up text and R results.
Important: This is a business course and, as such you are required to submit all exercises,
homework and project reports with a professional, businesslike appearance, free of
grammatical errors and typos, and with well articulated interpretation narratives. No
knitting, improper knitting and submissions with writing and formatting issues will
have up to 3-point (out of 10) deductions for exercises and up to 10-point (out of
100) deductions in homework.
Quarto contains three main types of content:
1. The YAML (YAML Ain’t Markup Language) header, which is where you place the
title, author, date, type of output, etc. It is at the top of the R Markdown file and
starts and ends with ---. I suggest using the format docx.

2. Markup sections, which is where you type any text you wish, which will show up as
typed text. You will learn these later.

3. Code chunks: which is where you write your R code. An R code chunk starts with a
```{r} and ends with a ```.

Your knitted file must:

• Display all your R commands (leave echo: true in the YAML. FYI, echo: false
suppresses the R code)
• Display the resulting R output results
• Contain any necessary text and explanations, as needed; and
• Be formatted for good readability and in a businesslike manner
• Be in the same order as the questions and with the corresponding question numbers

1. Basic R Concepts
1.1 Write a simple R function named area() that takes 2 values as parameters (x and y,
representing the two sides of a rectangle) and returns the product of the two values
(representing the rectangle’s area). Then use this function to display the area of a rectangle
of sides 6x4. Then, use the functions paste(), print() and area() to output this result:
The area of a rectangle of sides 6x4 is 24, where 24 is calculated with the area()
function you just created.
area <- function(x,y) {return(x*y)}
area(4,6)

[1] 24

print(paste("The area of a 4x6 rectanlge is", area(4,6)))

[1] "The area of a 4x6 rectanlge is 24"

1.2 Write a simple for loop for i from 1 to 10. In each loop cycle, compute the area of a
rectangle of sides i and i*2 (i.e., all rectangles have one side double the lenght than the
other) and for each of the 10 rectangles display “The area of an 1 x 2 rectangle is 2” for i=1,
“The area of an 2 x 4 rectangle is 8”, and so on.
for (i in 1:10) {
print(paste("The area of a", i, "x", i * 2,
"rectangle is", area(i, 2 * i)))
}

[1] "The area of a 1 x 2 rectangle is 2"

[1] "The area of a 2 x 4 rectangle is 8"
[1] "The area of a 3 x 6 rectangle is 18"
[1] "The area of a 4 x 8 rectangle is 32"
[1] "The area of a 5 x 10 rectangle is 50"
[1] "The area of a 6 x 12 rectangle is 72"
[1] "The area of a 7 x 14 rectangle is 98"
[1] "The area of a 8 x 16 rectangle is 128"
[1] "The area of a 9 x 18 rectangle is 162"
[1] "The area of a 10 x 20 rectangle is 200"

2. Data Manipulation
2.1 Copy the Credit.csv data file to your working directory (if you haven’t done this yet).
Then read the Credit.csv data file into a data frame object named Credit (Tip: use the
read.table() function with the parameters header=T, sep=",", row.names=1). Then, list
the first 5 columns of the top 5 rows (Tip: use Credit[1:5, 1:5])
Credit <- read.table("Credit.csv",
header = T,
sep = ",",
row.names = 1)
Credit[1:5, 1:5]

Income Limit Rating Cards Age

1 14.891 3606 283 2 34
2 106.025 6645 483 3 82
3 104.593 7075 514 4 71
4 148.924 9504 681 3 36
5 55.882 4897 357 2 68

2.2 Using the class() function, display the object class for the Credit data set, and for
Gender (i.e., Credit$Gender), Income and Cards
class(Credit)

[1] "data.frame"

class(Credit$Gender)

[1] "character"
class(Credit$Income)

[1] "numeric"

class(Credit$Cards)

[1] "integer"

2.3 Create a vector named income.vect with data from the Income column. Then use the
head() function to display the first 6 values of this vector.
income.vect <- Credit$Income
head(income.vect)

[1] 14.891 106.025 104.593 148.924 55.882 80.180

3. Basic Descriptive Analytics

3.1 Compute the mean, minimum, maximum, standard deviation and variance for all
the values in the income.vect vector. Store the respective results in variables name
mean.inc, min.inc, etc. Then, use the c() function to create a vector called income.stats
with 5 values you computed above. Then use the names() function to give the
corresponding names “Mean”, “Min”, “Max”, “StDev”, and “Var”. Then display the
income.stats vector, but wrap it within the round() function with a parameter digits = 2
to display only 2 decimals.
Technical Note: The names() function needs to create a vector with the respective names
above, which need to correspond to the values in income.vect. Therefore, you need to use
the c() function to create a vector with these 5 names.
mean.inc <- mean(income.vect)
min.inc <- min(income.vect)
max.inc <- max(income.vect)
sd.inc <- sd(income.vect)
var.inc <- var(income.vect)

income.stats <- c(mean.inc, min.inc, max.inc, sd.inc, var.inc)

names(income.stats) <- c("Mean","Min","Max","StDev", "Var")
round(income.stats, digits = 2)

Mean Min Max StDev Var

45.22 10.35 186.63 35.24 1242.16

3.2 Display a boxplot for the predictor Income. Tip: you can do this 2 ways. First you can
attach() the Credit data set (which loads the data set in the work environment) and then
do a boxplot() for Income. Or, do it without attaching, but using the table prefix (i.e.,
Credit$Income). Use the xlab = attribute to name include the label “Income”. Then display
similar boxplots but this time broken down by Gender (i.e., Credit$Income ~
Credit$Gender).
boxplot(Credit$Income,
xlab = "Income")

boxplot(Credit$Income ~ Credit$Gender)

3.3 Display a histogram for the variable Rating, with the main title “Credit Rating
Histogram” (main =) and X label “Rating” (xlab =). Then draw a QQ Plot for Rating (Tip:
use the qqnorm() function first to draw the data points and then use the qqline() function
to layer the QQ Line on top).
hist(Credit$Rating,
main = "Credit Rating Histogram",
xlab="Rating")

qqnorm(Credit$Rating)
qqline(Credit$Rating)
3.4 Briefly answer in your own words: Do you think that this data is somewhat normally
distributed? Why or why not? In your answer, please refer to both, the Histogram and the
QQ Plot.
# The data is somewhat normal in the middle, but the qqplot deviates from the
qqline providing some indication of non-normality at the tails. The histogram
shows some skewness to the right indicating some departure from normality, bu
t it has a bell shape in the center of the data, which is consistent with the
QQ Plot.

4. Basic Predictive Analytics

4.1 First, enter the command options(scipen = 4) to minimize the display values with
scientific notation. Then, create a simple linear regression model object with the lm()
function to fit credit Rating as a function of Income and save the results in an object
named lm.rating. Then display the model summary results with the summary() function.
Tip: use the formula Rating ~ Income, data = Credit inside the lm() function.
options(scipen = 4)
lm.rating <- lm(Rating ~ Income,
data = Credit)
summary(lm.rating)

Call:
lm(formula = Rating ~ Income, data = Credit)

Residuals:
Min 1Q Median 3Q Max
-173.855 -79.417 -0.384 79.747 171.955

Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 197.8411 7.7089 25.66 <2e-16 ***
Income 3.4742 0.1345 25.83 <2e-16 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

Residual standard error: 94.71 on 398 degrees of freedom

Multiple R-squared: 0.6263, Adjusted R-squared: 0.6253
F-statistic: 667 on 1 and 398 DF, p-value: < 2.2e-16

4.2 Now, plot Credit Rating (Y axis) against Income (X axis), with respective labels “Income”
and “Credit Rating”. Tip: feed the same formula you used in the lm() function above, but
using the plot() function instead. Then draw a regression line by feeding lm.rating into
the abline() function.
Note: notice that I added the parameters #| fig-width: 8 and #| fig-height = 8 to
control the size of the figure. Notice that the YAML has fig-width: 10 and fig-height: 6,
which are global parameters affecting the entire document. You can change any parameters
withing a code cell as I did below to override a global parameter, just for the specific code
cell. The rest of the script is unaffected.
plot(Rating ~ Income, data = Credit)
abline(lm.rating)

4.3 Write a simple linear model to predict credit ratings using these predictors: Income,
Limit, Cards, Married and Balance. Name the resulting model lm.rating.5. Then display
the regression using the summary() function.
lm.rating.5 <- lm(Rating ~ Income + Limit + Cards + Married + Balance,
data = Credit)
summary(lm.rating.5)

Call:
lm(formula = Rating ~ Income + Limit + Cards + Married + Balance,
 data = Credit)

Residuals:
Min 1Q Median 3Q Max
-24.0051 -7.0024 -0.9291 6.3789 26.2751

Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 27.1070066 2.1867611 12.396 < 2e-16 ***
Income 0.0975008 0.0335195 2.909 0.00383 **
Limit 0.0641536 0.0009004 71.247 < 2e-16 ***
Cards 4.7108256 0.3762419 12.521 < 2e-16 ***
MarriedYes 2.1217503 1.0441007 2.032 0.04281 *
Balance 0.0084355 0.0031308 2.694 0.00735 **
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

Residual standard error: 10.14 on 394 degrees of freedom

Multiple R-squared: 0.9958, Adjusted R-squared: 0.9957
F-statistic: 1.85e+04 on 5 and 394 DF, p-value: < 2.2e-16

4.4 Question: what do you think are the most influential predictors of credit rating?
# All predictors are statistically significant (i.e., they have asterisks nex
t to them and the p-values are smaller than 0.05). Also, all predictors are p
ositive, so they all have a positive influence on credit rating. Limit and Ca
rds are the most significant and the number of Cards seems to have the strong
est effect.