| Course Title | Probability and Statistics for Computer Science |
|---|---|
| Time | Wednesday, 6:30 PM - 9:00 PM |
| Location | NA 6311 |
| Credits & Hours | 3 Credits, 3 Hours |
| Instructor | Evan Agovino |
| evan.agovino.ccny@gmail.com | |
| Office | Varies |
| Office Hours | Varies, by appointment |
This course is an introduction to the practical tools of probability and statistics, including but not limited to descriptive statistics, probability theory, knowledge of discrete and continuous distributions, random variables and estimation, hypothesis testing and regression. The course takes a computational and applied approach to these topics. Though we will cover the mathematical theory behind our topics, students' output will involve applying said theory to real-world problems and datasets.
The course will be presented entirely in Python to mimic the workflow and tools used by professional Data Analysts and Data Scientists. No prior knowledge of Python is required. The majority of the class will use a few core packages, including Numpy for number simulation and statistical learning, Pandas for data exploration and cleaning, and Matplotlib for data visualization. In-class assignments, homework, and projects will be submitted via Jupyter Notebook files.
Classtime will be split between lectures and hands-on group work, with occasional quizzes, announced and unannounced, to check for understanding. In-class participation is essential to the course as a means of understanding and applying the concepts covered in lecture.
Math 20100 with minimum C grade, CSC 10300, CSC 10400.
By the end of the course, students should be proficient at:
-
Single Variable Explorations: Examine a single variable, understand its underlying distribution, and choose the appropriate summary statistics for it.
-
Pair-Wise Exploration: Identify possible relationships between variables and compute correlations and linear fits.
-
Estimation and Hypothesis Testing: Understand the following three questions when reporting statistical results: 1) How big is the effect? 2) How much variability should we expect if we run the same measurement again? 3) Is it possible that the apparent effect is due to chance?
-
Visualization: Use data visualization as a tool for examining data and communicating results
| Weight | |
|---|---|
| Group Project | 25% |
| Midterm Exam | 25% |
| Final Exam | 25% |
| Homework/Quizzes | 15% |
| Participation | 10% |
Group Project: Students will work on a small group project throughout the second half of the course that they will be expected to present to the class at the end of the semester. Projects will be graded based on a demonstration of core principles taught in class, and effectiveness of communication in their presentation. Details of the project will be shared later in the semester.
Exams/Quizzes: The midterm and final exams will focus on the core concepts covered in the class and will mimic the style of questions frequently asked in interviews for data-related roles. In-class quizzes, both announced and unannounced, will occasionally be administered to check for understanding.
Homework: Students will be given weekly homework assignments in Python to check for comprehension of the material. Homework will be graded on a 5-point scale for completeness and effort. Homework will be due at the beginning of class every week. Any homework submitted after 6:30 PM on the Wednesday it is due will be scored as 0/5. Exceptions will be granted only as mandated by CUNY policy.
Participation: Students are expected to attend class and be active participants in discussion.
Students will be given weekly reading assignments from the following texts:
-
Introduction to Probability and Statistics for Engineers and Scientists, Sheldon M. Ross, Third Edition. Available here
-
Think Stats: Exploratory Data Analysis in Python, Allen B. Downey, Second Edition. Available here
Students will be given additional readings throughout the semester related to the material on a given week. Additional readings will be shared in the Github repo as they are added.
All class materials, including the syllabus, readings, assignments and more can be accessed at the Github repository here.
Students will use Binder as a means of running Python via Jupyter Notebook. Binder is a cloud-based executable environment that lets anyone interact with a Jupyter notebook through an internet browser (and no registration required!). The Binder link is below and may be periodically updated throughout the semester.
Students will receive an invite to join a Slack channel for the class, which they are required to join. While email and/or Blackboard (TBD) may be used as a channel for some administrative updates, updates may also be sent via the Slack channel, which students are responsible for monitoring. Students are also encouraged to use Slack to communicate with each other to work on class materials, replicating the workflow used by professionals.
| Week | Date | Topic |
|---|---|---|
| 1 | January 30 | Introduction |
| 2 | February 6 | Descriptive Statistics |
| 3 | February 13 | Basic Probability I |
| 4 | February 27 | Basic Probability II |
| 5 | March 6 | Random Variables and Distributions |
| 6 | March 13 | The Normal Distribution and Central Limit Theorem |
| 7 | March 20 | Midterm |
| 8 | March 27 | Estimation and Confidence Intervals |
| 9 | April 3 | Hypothesis Testing |
| 10 | April 10 | Relationships Between Variables |
| 11 | April 24 | Regression |
| 12 | May 1 | Regression II |
| 14 | May 8 | Project Presentation |
| 15 | May 22* | Final Exam |
There will be no class on Wednesday, April 17 due to Spring Recess and no class on Wednesday, May 15 due to Reading Day.
*Final Exam date subject to change
The CUNY Policy on Academic Integrity. The policy, as adopted by the Board, is available to all students. Academic dishonesty is prohibited in the City University of New York and is punishable by penalties, including failing grades, suspension, and expulsion.