With the recent advances in powerful computing and the availability of massive sets of data, the tools of statistics, data science, and analytics have become indispensable in the applied sciences and in industry. This course focuses on the mathematical underpinnings that provide the foundations to modern day data analysis and statistics.

  • Parametric models, linear models, variable selection, the lasso and matrix completion
  • Estimation, criteria and construction of estimators, maximum likelihood, asymptotics
  • Non parametric models, empirical distribution function, bootstrap
  • Hypothesis testing, multiple hypotheses testing, family wise error, false discovery rate
  • Density and regression estimation, regularization and smoothing
  • Reproducing Kernel Hilbert Space methods

Course Prerequisite: Students should have at least one good course in probability, and some basic statistics. It will be assumed that students are familiar with the first five chapters of the course text, All of Statistics: A concise course in Statistical Inference, by Larry Wasserman. Students should review these chapters and study any material new to them before starting the course.

It is also strongly recommended that students read Chapter 6 of the textbook, which consists mostly of material that is covered in first year statistics courses (e.g., confidence intervals, testing hypotheses).

Instructor: Larry Goldstein,  Office hours Mondays and Thursdays, 4:00-5:10, Room 610

Feedback Form

Structure and Evaluation

Course participation, 25%, Midterm Exam , Monday August 3rd, 17:15, 30% : 22, 27, 29, 35, 43, 55, 60, 108

Final Exam, 45%. Thursday 13/08, 16,15 – 17:45, Meeting room – V floor

Course Text
All of Statistics: A concise course in Statistical Inference, by Larry Wasserman.

Additional References

Assignments:

Assignment #1

Assignment #2

Assignment #3

Assignment #4

Assignment #5

A Selection of Works by the Instructor:

Gaussian Phase Transitions and Conic Intrinsic Volumes: Steining the Steiner Formula
Goldstein, L., Nourdin, I. and Peccati, G.
Annals of Applied Probability (2017), vol 27, pp. 1-47
[http://arxiv.org/abs/1411.6265]

Relaxing the Gaussian assumption in Shrinkage and SURE in high dimension
Fathi, M., Goldstein, L., Reinert, G. and Saumard, A.
Annals of Statistics, (2022) vol 50, No 4, 2732–2766
[https://doi.org/10.1214/22-AOS2208]

Non-Gaussian Observations in Nonlinear Compressed Sensing via Stein Discrepancies
Goldstein, L. and Wei, X
Information and Inference: A Journal of the IMA, (2019) vol 8.1, pp. 125-159. iay006.
[https://arxiv.org/abs/1609.08512]

M-estimation in a diffusion model with application to biosensor transdermal blood alcohol monitoring
Allayioti, M., Bartroff, J., Goldstein, L., Luczak, S. and Rosen, G.
[https://arxiv.org/abs/2002.05335]

Gaussian random field approximation via Stein’s method with applications to wide random neural networks
Balasubramanian, K., Goldstein, L., Ross, N. and Salim, A.
Applied and Computational Harmonic Analysis 72 (2024): 101668.
[https://arxiv.org/abs/2306.16308]

Optimal Plug-in Estimators for Nonparametric Functional Estimation
Goldstein, L. and Messer, K.
Annals of Statistics, (1992) vol. 20, No. 3, 1306–1328
[https://doi.org/10.1214/aos/1176348770]

 

Other Links of Interest