RTMNU · Final Year B.Tech Computer Engineering · CBCS
Predictive Analytics: Regression and Classification
BTCME802.1T · Elective VII · Semester VIII · 3 credits · College 30 + university 70
One of the Elective VII options. An NPTEL course: its page lists twelve weeks, grouped here into five units.
Unit 1: Least Squares and the Regression Model (weeks 1–3)
The landscape of predictive models; the least-squares method, normal equations and the Gauss–Markov theorem; the geometry of the regression model, feature engineering and statistical inference on regression coefficients.
Examined in the 70-mark end-semester paper
- Least squares estimation and the normal equations
- Gauss–Markov theorem
- Inference on regression coefficients
- Multiple correlation and multiple regression
Unit 2: Model Assumptions, Comparison and Complexity (weeks 4–6)
Checking model assumptions; comparing models with R-squared, RMSE, AIC or BIC; model complexity and the bias–variance trade-off, feature selection and dimension reduction; multicollinearity and the variance inflation factor; regularisation with LASSO, ridge and elastic net.
Examined in the 70-mark end-semester paper
- Regression diagnostics and model assumptions
- Model comparison: R-squared, RMSE, AIC and BIC
- Multicollinearity and the variance inflation factor
- Regression basics and the bias–variance trade-off
- Feature selection: sequential forward and backward
- Regularisation: ridge and lasso
Unit 3: Regression in Practice and its Applications (weeks 7–8)
Regression analysis with Python, R and Julia; applications — the capital asset pricing model, bootstrap regression, time-series forecasting with regression and the Granger causal model.
Examined in the 70-mark end-semester paper
- Applications of regression: CAPM, bootstrap and time-series forecasting
Unit 4: Logistic Regression and Discriminant Analysis (weeks 9–11)
Logistic regression and maximum-likelihood estimation of its coefficients, fitted in R, Python and Julia; inference for logistic regression; discriminant analysis.
Examined in the 70-mark end-semester paper
- Logistic regression
- Maximum likelihood and method of moments
- Discriminant functions
Unit 5: Generalised Linear Models (week 12)
Multinomial logit regression, generalised linear regression, and Poisson and negative binomial regression.
Examined in the 70-mark end-semester paper
- Multinomial logistic regression
- Generalised linear models: Poisson and negative binomial regression
Unit contents are summarised from the official RTMNU syllabus . AiCanCode.org is not affiliated with Rashtrasant Tukadoji Maharaj Nagpur University (RTMNU).