Concept

Stepwise regression

Related lectures (31)

LASSO Regression: Sparse Signal Induction

Explores LASSO regression for inducing sparsity in signals through gradient descent.

Model Selection Methods in Biostatistics

Explores model selection methods in biostatistics, emphasizing the importance of starting with a sensible model.

Compares L1 and L0 penalization in linear regression with orthogonal designs using greedy algorithms and empirical comparisons.

Model Selection: Least Squares

Explores model selection in least squares regression, addressing multicollinearity challenges and introducing shrinkage techniques.

Data-Driven Modeling: Regression

Introduces data-driven modeling with a focus on regression, covering linear regression, risks of inductive reasoning, PCA, and ridge regression.

Model Building: Linear Regression

Explores model building in linear regression, covering techniques like stepwise regression and ridge regression to address multicollinearity.

PhD Programs at EPFL

Covers PhD programs at EPFL, including selection procedures, research areas, and support for students.

Multivariate Time Series: Cointegration & Forecasting

Explores multivariate time series analysis, cointegration, forecasting with ARMA models, and practical applications in interest rates analysis.

Variational Inference and Neural Networks

Covers variational inference and neural networks for classification tasks.

Geometry of the Lasso

Explores the geometric explanation of why Lasso solutions are sparse and how coefficients change with the regularization parameter.

Variable Selection Methods

Explores variable selection methods and the bias-variance trade-off in statistical modeling.

Polynomial Regression and Gradient Descent

Covers polynomial regression, gradient descent, overfitting, underfitting, regularization, and feature scaling in optimization algorithms.

Robust Regression: Methods and Applications

Explores robust and resistant methods in linear models, emphasizing the importance of handling extreme observations and the implications of robustness in regression models.

Regularization in Machine Learning

Explores Ridge and Lasso Regression for regularization in machine learning models, emphasizing hyperparameter tuning and visualization of parameter coefficients.

Model Evaluation and Selection

Discusses model evaluation, cross-validation, and variable selection in machine learning.

Regression Trees and Ensemble Methods in Machine Learning

Discusses regression trees, ensemble methods, and their applications in predicting used car prices and stock returns.

Variable Selection Methods: Subset vs. Container Approaches

Explores variable selection methods in machine learning, including subset and container approaches, using exhaustive and greedy procedures.

Specification Testing and Machine Learning

Explores specification testing, machine learning, overfitting, regularization, prediction tests, and variable selection.

Testing and Debugging

Explores the history of computing, automated testing, TDD, test coverage, bug reports, and debugging.

Singular Value Decomposition: Ridge and LASSO

Covers the concept of Singular Value Decomposition and its application in regression techniques.