COSC 6323: Statistical Methods

Teaching

To fulfill its mission, AC-DC carries out challenging human studies, which accrue big data, requiring sophisticated analysis. The skills required for such studies is a superset of the skills required for any other data-driven project.

END OF SEMESTER COMMENTS

★ ★ ★ ★ ★
Everything is super good Professor has really good grip on what he is delivering and main thing i want mention is about the teaching assistant for this course Kiran Fetah is superb in helping students in all the way they need .
Thanks

★ ★ ★ ★ ★
Professor and TA's gave continuous support. I am gonna really miss this class.

★ ★ ★ ★ ★
The most positive thing is conducting TA office hours which is very useful to clarify doubts for solving assignments There is nothing like negative but the deadline time for assignments and project is bit hectic when it comes to the end of the semester though professor and TA worked a lot to give us more time

★ ★ ★ ★ ★
Both Professor and TA's are very supportive and helpful.

★ ★ ★ ★ ★
The topics were explained very clearly and thanks to professor and TA's help, we were able to grasp and solve problems statistically. They are no negative attributes of this class.

★ ★ ★ ★ ★
Too many assignments made me feel a little stress

★ ★ ★ ★ ★
lot of pratical knowledge learnt. overall good subject and good teaching

★ ★ ★ ★ ★
The positive is Professor way of teaching is at top level but coming to the assignments there was a ambiguity between the professor and TA's explanation

★ ★ ★ ★ ★
I really liked the weekly assignment that allow us to apply the concepts we have learnt in class, I like how professor deliver the entire concepts in class using sceneriors and thing we could very much relate to in everyday life. I very much appreciate the effort of the T.A's in ensuring we get the concept and apply them rightly in solving real word problems through their numerous office hours and rapid responses to personal messages on teams. But sometime the assignments get overwhelming.

★ ★ ★ ★ ★
The good thing is that this gave a good exposure to the applications on real world problems. There is nothing much negative but the only thing is the assignments sometimes are challenging. Overall, I feel very glad that I took this subject.

★ ★ ★ ★ ★
I found it awesome about lectures and TA hours
It may be some what fair if you can decrease the workload and reduce the difficulty of assignments
Also,if you can consider 10 marks as one grade, i would be satisfied in terms of Results.

★ ★ ★ ★ ★
Overall a good learning experience Thanks

★ ★ ★ ★ ★
Positive: Everything goes on in a sequential pattern. Lecture followed by TAs practical session, assignment, Office hours. The grading was on point and punctual. The TAs were very supportive. Negative: The course load was more because of assignments every week. But, that has helped us in learning alot.

★ ★ ★ ★ ★
Positive Attributes: Helps in the way to make sense of the complex data and to make decision. Learned about the assumptions that underline statistical models and the importance of understanding the limitations of these models. Negative Attributes: Difficulty in communicating the results.

★ ★ ★ ★ ★
I thoroughly enjoyed the course and found it to be an enriching learning experience. The course content was well-structured, and the topics covered provided a comprehensive understanding of the subject matter. Throughout the course, I gained valuable insights into statistical methods and their application in research. The practical examples and case studies helped in reinforcing the concepts and made the learning process engaging and enjoyable. I particularly appreciated the emphasis on real-world applications, which helped me relate the material to my own research interests.

★ ★ ★ ★ ★
Classes were informative and we had weekly assignments which was a bit stressful but managed to finish them on time. Each one was a new task and helped me learn new things. Our TA's were very helpful.

★ ★ ★ ★ ★
Negatives are more number of assignments.

★ ★ ★ ★ ★
I learned so much from the continuos weekly assessment and projects from real world data... This course will certainly help me a lot in my career !

★ ★ ★ ★ ★
Positive:

Very focused on real datasets
All assignments gave me tools and libraries I could directly use in my research.
The focus on practical details and statistical "tricks" over dry theory was very appreciated.
The labs were great, very in-depth.

Negative:

Homeworks/projects, while well-designed, required a lot of extra support from the TAs to be doable.
Solving HWs without attending office hours resulted in poor scores, which is annoying since all the requirements should be in the questions.

★ ★ ★ ★ ★
The course exceeded my expectations in terms of both content and instruction. The professor's, Vitalii's and Fettah's expertise, enthusiasm, and commitment to student success made this an exceptional learning experience, and I am grateful for the opportunity to have taken this course.

KEY INFORMATION

Instructors

Grading

Prof. Ioannis Pavlidis (This email address is being protected from spambots. You need JavaScript enabled to view it.) Office Hours: 3-4 pm on Fridays @ TEAMS
Vitalii Zhukov (This email address is being protected from spambots. You need JavaScript enabled to view it.) Office Hours: 11 am-12 pm on Mondays @ TEAMS
Fettah Kiran (This email address is being protected from spambots. You need JavaScript enabled to view it.) Office Hours: 10-11 am on Fridays @ TEAMS

13 x 3% Homework
61% Project

Day, Time and Room

Course Project

Friday, 4:00-7:00 pm @ TEAMS & HBS 315

Requirements

Class Video Resources

The project can be done either individually or in pairs.
Pairs need to be declared by the end of the second week of classes.

References

[1] Donna L. Mohr, William J. Wilson, Rudolf J. Freund. Statistical Methods. 4th Edition. Academic Press, 2021.

[2] Terence C. Mills. Applied Time Series Analysis. 1st Edition. Academic Press, 2019.

COURSE OUTLINE

Lesson 1: Statistics, Machine/Deep Learning, and Data Science 01/20/2023

Topics to Cover: Situating Statistics, Machine/Deep Learning, and Data Science; observations and variables; types of measurements for variables; distributions; numerical descriptive statistics; exploratory data analysis; bivariate data; data collection

Lesson 2: Probabilities and Sampling Distributions 01/27/2023

Topics to Cover: Probability; discrete probability distributions; continuous probability distributions; sampling distributions
Homework #1 due at 4 pm on 02/03/2023

Lesson 3: Principles of Inference 02/03/2023

Topics to Cover: Hypothesis testing; estimation; sample size; assumptions
Homework #2 due at 4 pm on 02/09/2023
Assignment of Projects

Lesson 4: Inferences on a Single Population 02/10/2023

Topics to Cover: Inferences on the population mean; inferences on a proportion; inferences on the variance; assumptions
Homework #3 due at 4 pm on 02/16/2023

Lesson 5: Inferences for Two Populations 02/17/2023

Topics to Cover: Inferences on the difference between means using independent samples; inferences on variances; inferences on means for dependent samples; inferences on proportions; assumptions and remedial methods
Homework #4 due at 4 pm on 02/23/2023

Lesson 6: Inferences for Two or More Populations 02/24/2023

Topics to Cover: Analysis of variance; linear model; assumptions; specific comparisons; random models; unequal sample sizes; analysis of means
Homework #5 due at 4 pm on 03/02/2023

Lesson 7: Linear Regression 03/03/2023

Topics to Cover: The regression model; estimation of parameters; inferences for regression; correlation; regression diagnostics
Homework #6 due at 4 pm on 03/09/2023

Lesson 8: Multiple Regression 03/10/2023

Topics to Cover: The multiple regression model; estimation of coefficients; inferential procedures; correlations; special models; multicollinearity; variable selection; detection of outliers
Homework #7 due at 4 pm on 03/23/2023

Lesson 9: Dummy/Interval Variable Models 03/24/2023

Topics to Cover: The dummy variable model; unbalanced data; models with dummy and interval variables; weighted least squares; correlated errors
Homework #8 due at 4 pm on 03/30/2023

Lesson 10: Experimental Designs 03/31/2023

Topics to Cover: Two-factor factorial experiment; randomized block design; randomized blocks with sampling; repeated measures designs
Homework #9 due at 4 pm on 04/06/2023

Lesson 11: Categorical Data 04/07/2023

Topics to Cover: Hypothesis test for a multinomial population; goodness of fit using the 𝜒2 test; contingency tables; loglinear model
Homework #10 due at 4 pm on 04/13/2023

Lesson 12: Logistic and Multinomial Regression 04/14/2023

Topics to Cover: Logistic regression; multinomial regression
Homework #11 due at 4 pm on 04/20/2023

Lesson 13: Nonparametric Methods 04/21/2023

Topics to Cover: One sample; two independent samples; more than two samples; rank correlation; the bootstrap
Homework #12 due at 4 pm on 04/27/2023

Lesson 14: Time Series 04/28/2023

Topics to Cover: Time series and their features; stationary processes (ARMA); nonstationary processes (ARIMA)
Homework #13 due at 4 pm on 05/04/2023

Lesson 15: Project Presentations 05/05/2023

Project Reports due at 4 pm on 05/04/2023

WEEKLY GRADES AND STUDENT COMMENTS

Week 15 - May 05, 2023

Comments from students

Week 14 - April 28, 2023

Comments from students

★ ★ ★ ★ ★
Everything is good

Week 13 - April 21, 2023

Comments from students

Week 12 - April 14, 2023

Comments from students

Week 11 - April 07, 2023

Comments from students

Week 10 - March 31, 2023

Comments from students

Week 9 - March 24, 2023

Comments from students

Week 8 - March 10, 2023

Comments from students

Week 7 - March 03, 2023

Comments from students

★ ★ ★ ★ ★
All good

Week 6 - February 24, 2023

Comments from students

★ ★ ★ ★ ★
Everything is good. Thank you.

★ ★ ★ ★ ★
This HW04 was not good with explanation and was hard to implement. It took a lot of my time to finish that. I hope from next week more description will provide by you and make it a little easier. Moreover, we also have a project to do.

Week 5 - February 17, 2023

Comments from students

★ ★ ★ ★ ★
Thanks for the lecture

★ ★ ★ ★ ★
I have some comments on weekly homework as follow: 1- we need rubrics for every homeworks since I missed the same grades for hw1(3 feedback) and hw2 (one feedback)!! 2- we need more details of requirements weekly homework, for example (For the curated PP, HR-E4 and HR-AW signals construct Q-Q plots. The plots should be done for each participant and laid out in the two-page format you are familiar with from Q1 in HW 1. The aim is to provide a comparative and insightful account of signal normality to the analyst. State your observations and thoughts) The solution showed to us in class was in four pages before and After and in Q2 mentioned in two-page! 3- we need the final solution of weekly homework.
Thank you!

Week 4 - February 10, 2023

Comments from students

★ ★ ★ ★ ★
No techincal difficulties this week, which is appreciated.

★ ★ ★ ★ ★
Today's class has lot of useful content. Thanks to professor for all the information provided in today's class about assignments and one more chance to makeup for Homework1

★ ★ ★ ★ ★
Very difficult and time consuming homeworks which require strong knowledge of R programming. Takes an entire week to complete leaving no time for other classes.

Week 3 - February 03 , 2023

Comments from students

★ ★ ★ ★ ★
Unfortunately the online portion had lots of technical problems today, especially towards the end of the first session and continuing throughout Vitalii's demo. The audio would cut out for minutes at a time. I have no complaints about the content, just the stream itself.

★ ★ ★ ★ ★
The calss is intersenting, however I need time to figure out the homework, I sugget to have one week for each homework since I am taking two classes besides this class. Please.. I have one comment to Vitalii please explain slowly since I am beginner in R language. Professor and Vitali and Fettah are doing greet job.

★ ★ ★ ★ ★
I have zero knowledge in R programming and for someone like me I felt the homeworks are very difficult.

★ ★ ★ ★ ★
Dear professor, I had no experience in R language. The pace of class as I felt is fast but I'm covering it up by listening again the recorded lectures. The assignment, I had felt difficult to solve and moving forward since the deadline is fixed at Wednesday, I'm concerned about the course work. Kindly requesting you to be considerate while giving assignments.

★ ★ ★ ★ ★
Good

★ ★ ★ ★ ★
I felt Assignment was too tough. Firstly , I am new to R and as the fisrt assignment was like more than beginners level. So I have a request that if there is any chance of reducing the assignment level so that I can learn from foundations.

★ ★ ★ ★ ★
I felt Homework 1 was too difficult to complete even the deadline was extended because I am not very much aware of R. I kindly request you to decline the level of difficulty if you can so that I could learn from basics.

★ ★ ★ ★ ★
I feel each assignment should have a duration of 1 week to complete them as it is my first time working with R and it consumes a lot of time. And we also have other subjects to work on.

★ ★ ★ ★ ★
The assignment is so difficult it is like i am understanding the topic but when it comes to assignment it is not helping that much. And the assignment is taking a lot of time which is causing problem for my other courses.

★ ★ ★ ★ ★
Please give one week for every assignment

★ ★ ★ ★ ★
Found the assignment somewhat interesting but di...

★ ★ ★ ★ ★
I feel HW01 is very difficult. I have followed the class and office hours but it was so hard for me to get my head around the assignment. If this is the difficulty level for HW01, I wonder how the level of upcoming homeworks is going to be.

★ ★ ★ ★ ★
Hope to extend the work time.

★ ★ ★ ★ ★
The homework 1 was very very difficult and it is taking a lot of time to solve the questions and even the questions are not 100% understandable to do the task. It is even affecting my other courses as i need to spend a lot of time on this course.

★ ★ ★ ★ ★
I felt that Homework-1 was so difficult. As we are new to R we cannot code to that level at the begining itself.I request to reduce the difficulty as the deadline was also a mere 3 days.

Week 2 - January 28, 2023

Comments from students

★ ★ ★ ★ ★
Professor and TA explanation is pretty good with examples.

Week 1 - January 20, 2023

Comments from students

★ ★ ★ ★ ★
Hope there was more time to explain the examples where form R studio.

★ ★ ★ ★ ★
The professor and TA has taught the class really well. The hands on session is very informative, I'd like to name it 'Crash Course on R'

★ ★ ★ ★ ★
Very good and informative session

★ ★ ★ ★ ★
If time permits, I'd recommend a very short introduction to tidyverse, and all the verbs that come with it. Especially the %>% (pipe operator) and dplyr seem very in-use these days.

★ ★ ★ ★ ★
Good session overall

★ ★ ★ ★ ★
The class pace can be bit slowed down for better understanding.

★ ★ ★ ★ ★
As someone who has never used R, I found the intro a bit rushed. However for most it would likely be fine. As we are given the recording, it allows me to rewatch with pause to absorbe it. \nAlso, I like that you are doing this - obtaining feedback early in the class. My first impression is that this will be unlike any university course I have ever taken, in a good way. I'm excited to see all of how it unfolds but from the materials perspective but also from an academic implementation perspective. \nThank you.

★ ★ ★ ★ ★
The class was interesting

★ ★ ★ ★ ★
Classes are interactive.

★ ★ ★ ★ ★
looks good, but if pace can be a bit slow , i think i can catch up easily.

★ ★ ★ ★ ★
It is good and i am looking forward to learn R much more.

★ ★ ★ ★ ★
The class was good as I learned the basics of R language

★ ★ ★ ★ ★
Looking forward to have a great semester.

★ ★ ★ ★ ★
No, comments everything was good.

★ ★ ★ ★ ★
The explanation is quite good and it would be more interesting if it includes much more kinds of examples.

★ ★ ★ ★ ★
Class was very Useful. Thank you.