Software Tools for Data Analysis
STA 9750
Michael Weylandt
Week 5 – Tuesday 2026-09-29, Thursday 2026-10-01
Last Updated: 2026-09-29

STA 9750 Week 5

Today: In Class Presentations #01: Project Proposals + Optional Enrichment: Additional Review of R

These slides can be found online at:

https://michael-weylandt.com/STA9750/slides/slides05.html

Upcoming TODO

Upcoming student responsibilities - Tuesday Section:

Date Time Student Responsibility
2026-10-05 11:59pm ET Mini-Project Peer Feedback #00 Due
2026-10-06 6:00pm ET Pre-Assignment #05 Due
2026-10-13
Classes on Monday Schedule – Columbus Day Conversion Day
2026-10-16 11:59pm ET Mini-Project #01 Due
2026-10-20 6:00pm ET Pre-Assignment #06 Due
2026-10-26 11:59pm ET Mini-Project Peer Feedback #01 Due
2026-10-27 6:00pm ET Mid-Semester Check-In Slides Due

Upcoming TODO

Upcoming student responsibilities - Thursday Section:

Date Time Student Responsibility
2026-10-05 11:59pm ET Mini-Project Peer Feedback #00 Due
2026-10-08 6:00pm ET Pre-Assignment #05 Due
2026-10-15 6:00pm ET Pre-Assignment #06 Due
2026-10-16 11:59pm ET Mini-Project #01 Due
2026-10-22 6:00pm ET Mid-Semester Check-In Slides Due
2026-10-26 11:59pm ET Mini-Project Peer Feedback #01 Due
2026-10-29 6:00pm ET Pre-Assignment #07 Due
2026-10-29 11:59pm ET Mid-Semester Teammate Peer Evaluations Due

Other Important Dates

Other Important Dates:

Date Instructor Action
2026-10-06 Estimated Return of Peer Comments #00
2026-10-13 Estimated Return of Meta-Review Comments #00
2026-10-13 Classes on Monday Schedule – Columbus Day Conversion Day #01
2026-10-15 Mini-Project Released #02
2026-10-19 Mini-Project Peer Feedback Assigned #01
2026-10-27 Estimated Return of Peer Comments #01

Course Project Proposals

Today: Course project proposal presentations (Official Description)

  • 6 minute presentation
  • Key topics:
    • Animating Question
    • Team Roster
  • Also discuss: Possible specific questions, data sources, analytical plan, anticipated challenges

Most important: team names!

Previous: Rat Pack, Subway Surfers, Going for Gold, etc.

After Proposals

Extra Review of R Basics - 100% Optional

  • Variables, Vectors, Types, Control Flow

Mini-Project #00

MP#00 due last Friday - Course Infrastructure Set-Up

  • Setup RStudio Project
  • Create GitHub Account and Repo
  • Connect Local Computer to GitHub
  • Deploy GitHub Pages

I enjoyed getting to know (a bit more about) you!

Mini-Project #00

Mini-Project #00 peer feedback assigned

  • Due 2026-10-05
  • 3 peer feedback assignments per student
    • Give 3 comments, get 3 comments

Peer Feedback Instructions can be found online

Mini-Project #00

Interactive script automates this process:

source("https://michael-weylandt.com/STA9750/load_helpers.R")
mp_pf_perform(0, github="YOUR_GITHUB_ID")

Will request the secret code I gave you via Teams when you completed MP#00

Will ask you a series of questions and save your responses in a specifically formatted file

MP#00 Peer Feedback

Ungraded assignment, so simple feedback:

  • One strength, one weakness, one suggestion for improvement (3x)

After completion, upload bspf file to Brightspace (for privacy / confidentiality)

MP PF Cycle

Aims of Mini-Project Peer Feedback:

  • Learn to read and evaluate code
  • In analysis, rarely right and wrong; definitely better and worse
  • Learn tricks to improve your own site

“Good artists copy; great artists steal.” – Steve Jobs

Most coding is reading - most reading is reading your own old code

Course Support

Asynchronous Support: MS Teams

Synchronous Support: Office Hours

  • Tuesdays and Thursdays (Zoom) at 5pm

Pre-Assignments

No pre-assignment today (no lecture)

Pre-Assignments resume before next class

  • Day before class at 6:00pm
  • Available on course website + Brightspace after 9pm
  • Unlimited re-tries so make sure you get 30/30!

Proposal Presentations

Presentation Order

Tuesday
Presentation Order Team
1 Team 4 (T) (RA+CLA+DU+PS)
2 Team 10 (T) (MI)
3 Team 9 (T) (CO+WC+KC+VMS+JC+RJG)
4 Team 6 (T) (MS+WC+CW+DS+KL)
5 Team 7 (T) (JZ+ICR+JW+HG+CA+WG)
6 Team 3 (T) (ND+KK+AW+CO+NK)
7 Team 2 (T) (HS+NR+NT+ZY+DEH)
8 Team 5 (T) (MAT+KJ+LM+SV)
9 Team 8 (T) (MC+AVS+BB+LG)
10 Team 1 (T) (TG+FB+JML+RM+XW)

Presentation Order

Thursday
Presentation Order Team
1 Team 2 (R) (NY+KG+ZL+SRK+HV+RJ)
2 Team 9 (R) (FC+SB+FT)
3 Team 6 (R) (JF+ER+JL+LK)
4 Team 1 (R) (TJ+LF+PR+DC+AAA)
5 Team 4 (R) (KS+MA+OM+WK+NC+AR)
6 Team 5 (R) (MR+YZ+KSLL+SSS)
7 Team 8 (R) (PL)
8 Team 3 (R) (SA+NSAP+GW+AGG)
9 Team 7 (R) (DR+AMJ+CRP+KG+JAGP+FI)
10 Team 10 (R) (ZT)

Wrap-Up

Orientation

  • Team Formation ✅
  • Team Contracts ✅
  • Proposal Presentations ⬅️
  • Mid-Semester Check-Ins
  • Mid-Semester Peer Evaluations
  • Final Presentations
  • Final Reports
  • Final Peer Evaluations

Next Time

  • Communicating Results (quarto) ✅
  • R Basics ✅
  • Data Manipulation in R ⬅️
    • Single-Table Data Manipulation ✅
    • Multi-Table Data Manipulation ⬅️
  • Data Visualization in R
  • Getting Data into R
  • Statistical Modeling in R

Joins and Pivots

  • Combining multiple tidy tables into one ‘analysis-ready’ data set
  • Rearranging data for ease of use

Upcoming Work

Upcoming work from course calendar

Reminder: Stick around for optional review

Upcoming Work

Life Tip of the Week

Easy Duplicate Emails with Gmail

If you have a gmail account, you actually have infinite accounts:

email+tag@gmail.com gets forwarded to email@gmail.com

for any (reasonable) tag

Use this to:

  • Track who is reselling your email
  • Get access to ‘new account sales’ / ‘signup promotions’
    • Typically based on new email
  • Quickly sort and organize emails

Musical Treat

Optional Review: R Fundamentals

Values

Basic “things” in R (“scalars”):

  • Numeric values: 10, 3.14, 0.0002, 1.234e5, 3 + 4i
    • R distinguishes between integer and numeric / double but you don’t need to
  • Character values: "Baruch", "Michael Weylandt",
    • Arbitrary length - matched quotes (double or single)
    • Mix quotes to put quotes inside a string: "He said to me: 'Code is great!'"
  • Logical values: TRUE, FALSE (no quotes)

Values

Use the class() function to see types:

class(3)
[1] "numeric"
class(3.14159)
[1] "numeric"
class("I love R!")
[1] "character"

Variables and Assignment

We can assign a name to a value:

x <- 3

Now anywhere we use x, the value 3 automatically is introduced:

x^2
[1] 9

Variables and Assignment

Variable names must be:

  • One word (no spaces)
  • All alpha, numeric, or underscores
  • Start with a letter

Avoid special “reserved words”: if, else, for, etc

Variables and Assignment

Assignment (<-) is the last operation, so can use to save results for later use

five_factorial <- 5 * 4 * 3 * 2 * 1
five_factorial
[1] 120

Then

six_factorial <- 6 * five_factorial
six_factorial
[1] 720

Vectors

An ordered collection of the same type is called a vector:

  • Create with c (“concatenate”):
x <- c(1, 2, 3)
class(x)
[1] "numeric"
length(x)
[1] 3

Vectors

Vectors are everywhere in R:

  • The “single” values we saw earlier are just vectors of length 1
length(3)
[1] 1

Vectors

Access specific elements of a vector with []:

x <- c("a", "b", "c")
x[2]
[1] "b"

Or give a vector of indices:

x[c(2, 1, 3)]
[1] "b" "a" "c"
x[c(2, 1, 2, 1, 3, 3)]
[1] "b" "a" "b" "a" "c" "c"

Vectors

Negative indices drop:

x[-2]
[1] "a" "c"

This is different from Python: not count backwards!

Functions

Functions take input (“arguments”) and produce results and side-effects:

x <- c(1, 4, 9, 16)
sqrt(x)
[1] 1 2 3 4

x is an input to the function sqrt

sqrt doesn’t “see” the name x; it sees the vector 1, 4, 9, 16

Not 100% true, but close enough!

Functions

Some examples:

  • Print to screen with some formatting: print
  • “Pure” print to screen (no formatting): cat
  • Combine strings: paste
  • Math: sin, sqrt
  • Load a package: library

Vectorization

Where possible, functions are vectorized:

x <- c(1, 2, 3)
y <- c(4, 5, 6)

x * y
[1]  4 10 18

Operations occur “in parallel” on matched elements

Control Flow

Two useful operations for small code snippets:

if(condition){
  do_if_true
} else { # Optional - can omit this 'side'
  do_if_false
}

This is a conditional operator:

  • Used to run code sometimes
    • Download a missing file, but not a file already present
    • Throw an error if something bad happens
    • Turn off parts of code with (if(FALSE){})
  • Everything between { and } is handled
  • else branch is optional

Compound Conditions

Can do ‘compound’ or nested if/else:

if(x > 10){
  cat("x is very positive.")
} else if(x > 0) {
  cat("x is a little positive.\n")
} else {
  cat("x is negative.\n")
}

Control Flow

Two useful operations for small code snippets:

for(element in vector){
  process_one_at_a_time(element)
}

Goes through vector taking out one element at a time:

nums <- c(1, 2, 3, 4, 5)
for(n in nums){
  cat(n, "squared is", n^2, "\n")
}
1 squared is 1 
2 squared is 4 
3 squared is 9 
4 squared is 16 
5 squared is 25

Don’t focus on these too much - better alternatives coming soon!

Control Flow

By default, the last line of a function is the returned value:

my_absolute_value <- function(x){
  if(x > 0){
    x
  } else {
    -x
  }
}

my_absolute_value(-3)
[1] 3
my_absolute_value(3)
[1] 3

Control Flow

Can override with return statement - instantly returns and ‘exits’ function:

say_hello <- function(name, scream=FALSE, quiet=FALSE){
  text <- paste("Hello", name)
  
  if(scream){
    text <- paste(toupper(text), "!!!")
  }
  
  if(quiet){return(text)} # Stop here if quiet and skip print
  
  print(text)
}
say_hello("Michael")
[1] "Hello Michael"
say_hello("Michael", scream=TRUE)
[1] "HELLO MICHAEL !!!"
say_hello("Michael", quiet=TRUE)
[1] "Hello Michael"

Control Flow

Overly complicated \(\text{sign}(x)\) function:

sign <- function(x){
  if(x > 0){
    1
  } else {
    if(x < 0){
      -1
    } else {
      0
    }
  }
}

Control Flow

Somewhat better \(\text{sign}(x)\) function:

sign <- function(x){
  if(x > 0){
    1
  } else if(x < 0){
      -1
  } else {
      0
  }
}

Control Flow

Decent \(\text{sign}(x)\) function:

sign <- function(x){
  if(x > 0) return(1) # Use 'return' to stop function here
  if(x < 0) return(-1)
  return(0)
}

Would be even better to vectorize

Exercises

Programming exercises to practice these concepts

Exercise #01

Q: Write a function f that does the following:

f(c(1, 2, 3))
The vector has 3 elements and is of type numeric
f(c("a", "b", "c"))
The vector has 3 elements and is of type character
f(1:5)
The vector has 5 elements and is of type integer

Hint: Use the cat function to print to screen.

f <- function(x){
  cat("The vector has", length(x), "elements and is of type", class(x))
}

Exercise #02

Q: Write a vectorized function to tell if numbers are even.

is_even(3)
[1] FALSE
is_even(c(3, 4, 5))
[1] FALSE  TRUE FALSE
is_even(1:10)
 [1] FALSE  TRUE FALSE  TRUE FALSE  TRUE FALSE  TRUE FALSE  TRUE

Hint: Use the %% operator to get remainders from division

is_even <- function(x)  (x %% 2) == 0

Exercise #03

Q: Write a function to count the even elements of a vector:

count_even(3)
[1] 0
count_even(c(3, 4, 5))
[1] 1
count_even(1:10)
[1] 5

Hint: Combine Q2 with the “sum of logical = count” trick.

count_even <- function(x)  sum(is_even(x))

Exercise #04

Q: The seq() function lets us construct sequences. What is the average (mean) of the first 23 odd numbers?

Hint: Read the documentation for ?seq before trying this question

mean(seq(from=1, by=2, length.out=23))
[1] 23

Exercise #05

Q: The alternating harmonic series is defined as:

\[1 - \frac{1}{2} + \frac{1}{3} - \frac{1}{4} + \frac{1}{5} + \dots\]

Show that this series converges to \(\ln(2)\) by taking a partial sum of the first one million elements.

Hint: Use recycling to get the signs right:

seq(1, 5) * c(1, -1)
[1]  1 -2  3 -4  5
sum(1/seq(1, 1e6) * c(1, -1))
[1] 0.6931467

Compare to:

log(2)
[1] 0.6931472

Exercise #06

Q: Write a function that computes the mean of a vector (Don’t use the built-in mean function)

Hint: Use the sum and length functions

my_mean <- function(x){
  sum(x) / length(x)
}

my_mean(8)
[1] 8
my_mean(1:10)
[1] 5.5

Exercise #07

Q: Write a vectorized function that returns \(\sqrt{x}\) if \(x\) is positive and 0 otherwise.

pos_sqrt(c(-4, -1, 1, 4, 9, -9))
[1] 0 0 1 2 3 0

Hint: The pmax (“parallel max”) function may be useful here:

x <- c(1, 3, 5,  7)
y <- c(5, 3, 10, 1)
pmax(x, y)
[1]  5  3 10  7
pos_sqrt <- function(x) sqrt(pmax(0, x))

Alternative - use the ifelse function for vectorized conditionals (but this has a warning because it tries to do both conditions)

Exercise #08

Q: Write a function that takes in a vector of characters and returns the longest.

long_string(c("a", "bc", "def"))
[1] "def"
long_string(c("My", "name", "is", "Michael"))
[1] "Michael"

Hint: The nchar and which.max functions may be helpful here.

long_string <- function(x)  x[which.max(nchar(x))]

Exercise #09

Q: Write a function that computes the factorial of a value. (Don’t use the built-in factorial function)

Hint: Use the seq and prod functions

my_factorial <- function(n){
  return(prod(seq(1, n)))
}

my_factorial(8)
[1] 40320

Exercise #10

Q: Write a function that computes the factorial of a value using a loop (instead of the prod function).

my_factorial <- function(n){
  fact <- 1
  for(x in seq(1, n)){
    fact <- fact * x
  }
  return(fact)
}

my_factorial(8)
[1] 40320

Exercise #11

Q: Show that: \[\sum_{k=0}^{\infty} \frac{1}{k!} = e\]

Use the built-in factorial function since it is vectorized.

sum(1/factorial(seq(0, 10000)))
[1] 2.718282

Exercise #12

Q: Write a function that takes a vector and returns the maximum element. (Don’t use the built-in max function.)

Hint: You will need to use a loop and a conditional here.

Hint: We started factorial at 1; start max at -Inf (why?)

my_max <- function(x){
  max_val <- -Inf
  for(v in x){
    if(v > max_val) max_val <- v
  }
  return(max_val)
}

my_max(c(1, -1, 3, -4, 10, -3))
[1] 10