MIS 306 Data Analysis: Forecasting [Spring 2025]
I. Ozkan
Aims and Objectives
This course provides a comprehensive introduction to forecasting methodologies for time series data, focusing on the analysis and prediction of economic and financial datasets. By the end of the course, students will be equipped to analyze real-world time series data and generate accurate forecasts. The primary statistical programming tool utilized is R, along with its associated packages. The course content is dynamic, evolving annually to incorporate the latest advancements and applications in the field. This year’s topics include, but are not limited to:
Introduction
R
- R and RStudio Installation & Resources
- R Basics
- Dates in R
- Packages for TS
- Rmarkdown for HW Assignments
- An Example: Financial Time Series
- R and RStudio Installation & Resources
Time series Concept: Discrete-continuous time series, domains, probability-time-frequency domain
Time series and causality: This is in-class discussion
Exploratory Time Series analysis
- Graphical Methods
- Numerical Methods
- Graphical Methods
Time Series Decomposition
Smoothing
Univariate Time Series Models
- Stationary Time Series Models (AR, MA, ARMA Processes)
- Non-Stationary Time Series Models (ARIMA Processes)
- Stationary Time Series Models (AR, MA, ARMA Processes)
Course Schedule
Tentative Course Schedule
| Week | Date | Topic | Chapter | Chapter_URL |
|---|---|---|---|---|
| 1 | 2024-02-17 | Introduction to forecasting and R | 1. Getting started | https://OTexts.com/fpp3/intro.html |
| 2 | 2024-02-24 | Date and Time in R | R. Date and Time: Lecture Notes | Lecture Notes: To be distributed |
| 3 | 2024-03-02 | Time series graphics | 2. Time series graphics | https://OTexts.com/fpp3/graphics.html |
| 4 | 2024-03-09 | Time series decomposition | 3. Time series decomposition | https://OTexts.com/fpp3/decomposition.html |
| 5 | 2024-03-16 | The forecaster’s toolbox | 5. The forecaster’s toolbox | https://OTexts.com/fpp3/toolbox.html |
| 6 | 2024-03-23 | Exponential smoothing | 8. Exponential smoothing | https://OTexts.com/fpp3/expsmooth.html |
| 7 | 2024-03-30 | Exponential smoothing | 8. Exponential smoothing | https://OTexts.com/fpp3/expsmooth.html |
| 8 | 2024-04-06 | Midterm (Tentative) | Midterm | Midterm |
| 9 | 2024-04-13 | ARIMA models | 9. ARIMA models | https://OTexts.com/fpp3/arima.html |
| 10 | 2024-04-20 | ARIMA models | 9. ARIMA models | https://OTexts.com/fpp3/arima.html |
| 11 | 2024-04-27 | ARIMA models | 9. ARIMA models | https://OTexts.com/fpp3/arima.html |
| 12 | 2024-05-04 | Multiple regression and forecasting | 7. Time series regression models | https://OTexts.com/fpp3/regression.html |
| 13 | 2024-05-11 | Presentations | ||
| 14 | 2024-05-18 | Presentations |
References and Suggested Readings
Textbook
Hyndman, R. J., & Athanasopoulos, G. (2021). Forecasting: principles and practice. OTexts.
Suggested Books
Jonathan D. Cryer and Kung-Sik Chan, “Time Series Analysis with Applications in R”, second edition, Springer
Shumway, Robert H., Stoffer, David S, “Time Series Analysis and Its Applications, With R Examples”, third edition, Springer
Tsay, R. S. Analysis of financial time series 3rd Edition. John Wiley & Sons.
Lecture Notes
Data Sources
There are several time series data sources available. We will cover some of the packages that we can use to download time series.
Time Series Data Summaries
The first step is to understand the basic [statistical] properties of the data set at hand, time series data for this course. Chapter 2 and Chapter 4 of the textbook are good reads as a start.
Data Cleaning Preprocessing
Data rarely come clean. There may be a need for cleaning them. For example, time series data may contain outliers, missing values and errors. Understanding the properties of data and cleaning them probably the very first step in time series analysis.
Please read related part of Exploratory Time Series Analysis
Forecaster’s Toolbox and Exponential Smoothing
Some Simple approaches for forecasting a time series data may be helpful in some cases. Chapter 5 of the textbook introduces these approaches. Chapter 8 focuses on Exponential Smoothing (Note: We may skip some of those approaches)
Modelling Univariate Stochastic Time Series
Univariate stationary time series data widely modelled as AR, MA, ARMA. ARIMA models are used for the Non-stationary time series. Chapter 9 of the textbook discusses these modelling approaches.
Evaluation Criteria
- Assignments/Labs: 30%
- Midterm Exam: 20%
- Final Exam: 30%
- Presentation: 20%
- Attendance (Bonus): 10%
Policies
Attendance: Regular attendance is expected and will be rewarded.
Late Submissions: Assignments submitted late will incur a penalty unless prior approval is granted.
Academic Integrity:
Academic integrity is fundamental to the academic mission of the university. Acts of academic dishonesty, including but not limited to plagiarism, cheating, fabrication, or unauthorized collaboration, undermine the learning process and violate university policies.
Specific guidelines include:
Plagiarism: Using someone else’s work, ideas, or words without proper attribution is strictly prohibited. This includes copying and pasting from any source, paraphrasing without citation, or submitting another person’s work as your own.
Cheating: Unauthorized use of materials, devices, or information during exams or assignments, including sharing or receiving answers, is not allowed.
Fabrication: Falsifying or inventing data, citations, or research is a breach of academic integrity.
Collaboration: While collaboration on group assignments may be permitted, sharing answers or work on individual tasks is not acceptable unless explicitly authorized.
Consequences: Violations of academic integrity will be addressed following the university’s academic policies, potentially leading to penalties such as assignment failure, course failure, or further disciplinary actions.