Group A (Marks: 30)
(Related Course - STAT-405: Econometrics)
This course is planned to implement practical econometric-related problems and to facilitate research techniques in real situations, integrated with economics, social science, and life-related problems.
Course Objectives:
The objectives of this course are to:
- Design a scheme and develop the ability to keep lifelong research-oriented work,
- Apply time-related problems, especially simultaneous equation theory, in assessing the performance of a structure,
- Formulate econometric models,
- Estimate and test models.
Learning Outcomes:
After completing this course, students will be able to:
- Analyze multiple linear regression models,
- Specify assumptions, formulate, and estimate appropriate models,
- Perform different tests to test regression parameters,
- Detect autocorrelation, multicollinearity, heteroscedasticity problems in datasets,
- Estimate parameters in the presence of autocorrelation, multicollinearity, and heteroscedasticity problems in datasets,
- Model multivariate time series and estimate the parameters,
- Forecast multivariate time series,
- Interpret the results and test their statistical significance,
- Model simple applications using statistical software on a variety of datasets.
Contents:
- Estimation of parameters in multiple linear regression models, testing regression parameters using likelihood ratio test, Wald test, Lagrange multiplier test, tests for autocorrelation, multicollinearity, heteroscedasticity.
- Estimation of parameters and analysis of data in the presence of autocorrelation, multicollinearity, and heteroscedasticity.
- Modeling and forecasting multivariate time series.
Rationale:
Group B (Marks: 20)
(Related Course - STAT-408: Machine Learning)
This course is designed to apply the theory of machine learning and show how to apply the algorithms to different problems.
Course Objectives:
The objectives of this course are to:
- Introduce the students to the application of machine learning,
- Provide a broad survey of approaches and techniques in machine learning,
- Learn to apply supervised learning algorithms,
- Learn to apply unsupervised learning algorithms,
- Evaluate and improve a model,
- Solve problems using specific algorithms,
- Design and analyze machine learning experiments.
Learning Outcomes:
After completing this course, students will be able to:
- Understand modern notions in data analysis-oriented computing,
- Develop a deeper understanding of several major topics in machine learning,
- Learn algorithms and techniques and their applications,
- Analyze and handle large data sets,
- Develop basic skills necessary to pursue research in machine learning,
- Formulate machine learning problems corresponding to different applications using real-world data,
- Work on projects concerning machine learning.
Contents:
Implementation as Lab Programming: Might be in R, Python, or MATLAB.
Module 1: Introduction to Machine Learning: Applications of machine learning, supervised vs. unsupervised learning, Python libraries suitable for machine learning.
Module 2: Supervised learning: The Naïve Bayes classifier, linear regression, logistic regression, model evaluation methods, bias-variance tradeoff, support vector machines and kernels, k-nearest neighbor, decision trees, support vector machines, neural network, model evaluation.
Module 3: Unsupervised Learning: K-Means Clustering, Hierarchical Clustering, Density-Based Clustering.
Module 4: Projects.