Teaching
Office hours: Tuesdays. 14:00-15:00 Location: C.121b
Current (Fall 2024)
Advanced statistical analysis and model-based learning (formerly ‘Advanced Statistics for Data Science’)
Past
Reichman CS 3798: An Introduction to Information Theory (Fall 2023)
Reichman CS 3676: Advanced statistics for data science (Spring 2023)
Reichman CS 3141: Introduction to Machine Learning (Spring 2024, Spring 2023)
Reichman CS 3890: Information-theoretical analysis of neural language models
Class material including students’ presentations is available at https://github.com/alonkipnis/ITnLM (Fall 2022)
Reichman CS 3676: Advanced statistics for data science (Spring 2022)
| Lecture 1 | course overview, introduction to the linear model, examples |
| Lecture 2 | linear model and algebraic properties of least squares fit |
| Probability Review | random vectors, conditional distributions, multivariate normal |
| Lecture 3 | distributional properties of least squares fit |
| Lecture 4 | t-test, F-test, Gauss-Markov thm., intro. to statistical inference |
| Lecture 5 | one-sample testing, significance, confidence interval, power |
| Lecture 6 | two-sample testing, ANOVA I |
| Lecture 7 | computing least squares with SVD, ANOVA II, multiple comparisons |
| Lecture 8 | multiple comparisons and false-discovery rate, simple regression I |
| Lecture 9 | simple regression II, confidence bands, interplay between variables |
| Lecture 10 | variable selection I: stepwise methods and cross-validation |
| Lecture 11 | variable selection II: regularization, violations of assumptions |
| Lecture 12 | violations of assumptions, course summary, followups |
Stanford STATS 285: Massive computational experiments, painlessly (Spring 2020-2021)
with David Donoho and Masha Lofti
Stanford STATS 207: Introduction to time-series analysis (Fall 2020-2021)
| Lecture 1 | course outline, examples of time series data, models for time series data |
| Lecture 2 | sample autocorrelation and basic theoretical constructs |
| Lecture 3 | time series regression |
| Lecture 4 | trend models and Data Wrangling |
| Lecture 5-6 | ARMA/ARIMA Modeling I |
| Lecture 7 | ARMA/ARIMA Modeling II |
| Lecture 8 | ARIMA/SARIMA |
| Lecture 9 | regression with autocorrelated errors and lagged regression |
| Lecture 10 | efficient markets hypothesis and GARCH (guest lecture by David Donoho) |
| Lecture 11 | spectral Analysis I |
| Lecture 12 | spectral Analysis II |
| Lecture 13 | spectral regression and principal components |
| Lecture 14 | state-space modeling and the Kalman Filter |
| Lecture 15 | estimation of state-space models |
| Lecture 16 | dynamic linear models with switching |
| Lecture 17 | stochastic volatility, Bayesian analysis of state-space models |
| Lecture 18 | bootstrap reality check and technical trading rules (guest lecture by David Donoho) |
| Lecture 19 | prophet (guest lecture by Sean Taylor) |
| Lecture 20 | high-dimensional data, DeepAR, VEST |