Description
Regression Models: The Foundation of Predictive Analytics
Regression is the backbone of data science—from economics to epidemiology. This **Johns Hopkins course** dives deep into linear models, logistic regression, and inference, using real datasets to teach you not just *how* to fit a model, but *how to interpret it correctly*.
What You’ll Learn
- Linear regression—least squares, residuals, diagnostics
- Multivariable regression—confounding, adjustment, interaction terms
- Logistic regression—odds ratios, classification, ROC curves
- Poisson regression—for count data (e.g., website clicks, disease cases)
- Inference—confidence intervals, p-values, model selection
Real Datasets Used
- Galton’s height data
- Boston housing prices
- Medical trial outcomes
- Galaxy survey data
Who Should Take This?
- Biostatisticians and epidemiologists
- Economists and social scientists
- Data analysts needing rigorous modeling skills
- Students in the JHU Data Science Specialization
Why Regression in 2025?
Deep learning gets the headlines, but **90% of real-world business problems are solved with regression**—done right.
Ready to master the workhorse of data science? Enroll now.
