| Session no. | Segment 1 | Segment 2 | Segment 3 |
|---|---|---|---|
| 1 | Exploratory Data Analysis | Inferential Statistics 1: Central Limit Theorem, Confidence intervals (Z) and hypothesis testing (Z). Will also include a refresher on using the normal distribution as a prerequisite | Introduction to Machine Learning |
| 2 | Visualization with R for EDA 1 | Inferential Statistics 2: The t-distribution, t confidence intervals and hypothesis testing for means, difference b/w paired and independant means; Analysis Of Variance | Linear Regression 1: Simple linear regression, coefficients, confidence intervals, hypothesis testing |
| 3 | Visualization with R for EDA 2 | The binomial, geometric and poisson distributions as a prerequisite for inference on categorical variables | Linear Regression 2: R^2, Multiple linear regression, F statistic. Including/excluding variables |
| 4 | Data transformation with R 1 | Inferential Statistics 3: Inference for categorical variables, Confidence intervals, hypothesis testing single/two proportions. Chi-square test for independance/goodness of fit | Linear Regression 3: Qualitative predictors 2/more levels, dummy variables, interaction, non-linear transformations of predictors |
| 5 | **Single segment:** Getting our hands dirty with some messy data. A single project that'll make employers drool over your work | ||
| 6 | Data Wrangling with R 1 | Inferential Statistics 4: Power, Bootstrap | Linear Regression 4: issues faced in fitting the linear model. Lab |
| 7 | Data Wrangling with R 2: Rules of tidy data | Probabilities refresher for Bayesian statistics | Classification 1: Logistic regression, multiple logistic regression |
| 8 | Data Wrangling with R 3: Impact of data tidying in analysis/visualization | Bayesian statistics 1: Bayesian vs. frequentist definitions of probability, examples of inference in each | Logistic regression for over 2 response classes. Lab |
| 9 | Data Wrangling with R 4: Relational data, strings | Bayesian statistics 2: Bayesian inference | K-NN. Lab, comparison. |
| 10 | **Single segment:** R as a programming language, or CODE-CAINE. You know, *sniff *sniff. *aaahhhhhh | ||
| 11 | Reporting 1: R Markdown | Bayesian Statistics 3: Decision making | Model performance measure. Resampling methods |
| 12 | Reporting 2: Graphics makeover/cosmetics | Bayesian Statistics 4: Bayesian Regression | Model selection approaches |