Introductory tutorial on data exploration and statistical models
summary
Hands-on pipeline that loads four heart-disease files, concatenates them, and cleans data with pandas/StatsModels/scikit-learn in an Anaconda/Jupyter environment; it drops variables with >33% missing, imputes small gaps with mode, and reduces 920→741 observations for analysis. Univariate and bivariate checks find a Pearson r of −0.06 (p≈0.06) for age vs cholesterol, a significant chi-square linking chest pain to diagnosis while gender fails as a moderator, multiple regression gives R_squared 0.278, and logistic fits report odds ratios (female ≈3.7×); decision trees, random forests, lasso, and clustering are applied later.
AI-written summary · about
chapters (16)
- 0:00Introduction
- 0:45Why data exploration matters
- 3:12What we will cover in this session
- 5:00Data management in research
- 12:09Exploratory data analysis fundamentals
- 21:12Handling missing data
- 25:07Univariate analysis
- 32:29Two-variable visualization
- 39:41Inferential statistics intro
- 48:13ANOVA and post hoc testing
- 55:30Chi-square tests
- 1:07:00Testing moderator variables
- 1:11:18Linear regression building
- 1:17:07Regression diagnostics
- 1:20:19Logistic regression
- 1:23:12Summary
related talks
17mTransforming Medical Records Into a Heart Attack Prediction Model
01h 53mAnalyze your data in Python
01h 24mInpsect and try to interpret your scikit-learn machine-learning models
45mA Practical Introduction to Data Science
42mManualRivas
53mResponsible AI with SciPy
44mA Practical-ish Introduction to Data Science
15mInterpretable Machine Learning with R