Special Topics: Regression Analysis
Regression analysis techniques are probably the most widely used statistical techniques in the universe. Applications of such techniques are widespread in almost every industry possible including healthcare, finance, marketing, engineering to name a few. This course provides an overview of regression analysis. It focuses on building a greater understanding, theoretical underpinning, and statistical software (SAS & R) for applying the regression models and its generalizations. Using the bigger umbrella of generalized linear models, it also covers the topics like design of experiments, logistic and poisson regression. It focuses on all aspects of model building process using real life data and statistical software.
- Develop a deeper understanding of the linear regression model and its applications and limitations.
- Know how to start building a model, then diagnose it and finally update it.
- Develop a greater familiarity with a range of related regression techniques and methods through a diverse set of theoretical and applied tools.
- Learn how to carry out a task from start to finish using statistical software.
Lectures: The course will be taught in hybrid style. All the lectures will be delivered online every week in two parts for 13 weeks. SAS & R programs will be delivered on line and students are expected to learn how to run them on their own machine. There will be two in class exams where students need to show up in person.
Discussion & Online Office Hours: Every Monday 5:00 to 6:00. Students can join online or call to discuss and get help.
Prerequisite: One semester of Statistics and familiarity with running programs in own machine.
Software: Need to purchase SAS license from Rutgers and download R for free.
Grade Determination: Two in class exams 40% each and 1 Project 20%.
Book: Introduction to Linear Regression Analysis: Edition 5, by Douglas C. Montgomery, Elizabeth A. Peck, G. Geoffrey Vining. Publisher: John Wiley & Sons.
Part-1: Review of Probability distribution Theory
Part-2: Review of Statistical Inference – Estimation, Confidence Interval & Testing
Part-1: Introduction to SAS
Part-2: Introduction to R
Part-1: Introduction to Simple Linear Regression(SLR)
Part-2: Inference about SLR Parameters
Part-1: Review of Basic Vector and Matrix Algebra
Part-2: Introduction to Multiple Linear Regression
Part-1: Inference in MLR Model
Part-2: Model Evaluation
Part-1: Variable Selection and Model Building
Part-1: Influencial Points and Outlier detection
Part-2: Categorical Predictors and Transformation
Week-8: Meeting Day – Exam - 1
Mid Term Exam
Week-9: Spring Break
Part-1: Introduction to Generalized Linear Models
Part-2: Introduction Design of Experiments
Part-1: Interactions & higher order models
Part-2: Cross Over and Other designs
Part-1: Introduction to Logistic Regression
Part-2: Inference and Other issues in Logistic Regression
Part-1: Introduction to Poisson Regression
Part-2: Inference and Other issues in Poisson Regression
Part-1: Introduction to Time Series Models
Part-2: Inference in Time Series Models
Part-1: Application of Time Series in Finance
Part-2: Concluding Remarks and Final Projects
Week-17: Final Exam and Project Due.