Astronomy 8824 (Prof. Martini, Autumn 2019)
Astronomy 8824
Numerical and Statistical Methods in Astrophysics
Autumn Semester 2019
Prof. Paul Martini
TTh 8:50-10:20am
4054 McPherson Laboratory
[Contact
|
Syllabus
|
Evaluation
|
Books and References
|
Schedule and Readings
]
- Office: 4021 McPherson Lab
- Office Phone: 614-292-8632
- Office Hours: by appointment or whenever my door is open
- E-Mail: martini.10@osu.edu
The class syllabus and course outline are in this PDF File.
The course grade will be based on problem sets (70%), and class participation and presentations (30%). There will not be a final exam.
Here are the problem sets and due dates:
- Problem Set 1 Due Sept 3
- Problem Set 2 Due Sept 17 (Starter Notebook)
- Problem Set 3 Due Oct 1
- Problem Set 4 Due Oct 29 (Starter Notebook)
- Problem Set 5 Due Nov 12
- Problem Set 6 Due Nov 26
- Problem Set 7 Due Dec 6 (ps7data.tgz)
The Schedule and Readings section below includes the in-class presentations.
I recommend the following two books:
Numerical Recipes by Press, Teukolsky, Vetterling, and Flannery.
This book is a good introduction to many numerical methods, and is available for a number of programming languages (although unfortunately not python). It should be more useful for descriptions of numerical methods than for the routines. The 3rd edition has more information than the previous ones, although the other editions have the essentials.
Statistics, Data Mining, and Machine Learning in Astronomy by Ivezic, Connolly, VanderPlas, and Gray.
We will mostly cover the first five chapters, which are the chapters that focus on statistics. Please read David Weinberg's Reader's Guide to Chapters 3-5 before you start reading. The book uses python, and so is a good way to improve your knowledge of that language.
If you are not already conversant in python, I recommend that you also obtain an introduction to the python programming language. One geared toward scientific programming is likely to be most relevant. David Weinberg has recommended this one to past classes, and I think it is really good too:
Effective Computation in Physics: Field Guide to Research with Python by Scopatz and Huff.
Python References
Here are links to other python references:
- Demitri Muna's Introduction to Python Slides
- Thomas Robitaille's Python for Scientists
- The python module AstroML was started to accompany Statistics, Data Mining, and Machine Learning in Astronomy and contains numerous resources
These references are primarily aimed at scientists. There are also a seemingly exponentially increasing number of other guides and tutorials online.
The following are all of the Tuesdays and Thursdays in the Autumn Semester. If we have class on a given day, I will include the topic, reading(s), and any in class presentation for that day. If definitely will not meet, I will make that clear. And if I have not determined this yet, I will mark the date with "TBD."
August 20 (Tuesday): Numerical Integration
- Readings: David Weinberg's Numerical Notes 1: Numerical Integration. Numerical Recipes Chapter 4, focus on 4.0 to 4.4.
August 22 (Thursday): No Class - Department Symposium
August 27 (Tuesday): No Class
August 29 (Thursday): No Class
September 3 (Tuesday): Finish Numerical Integration, start Ordinary Differential Equations
- Readings: David Weinberg's Numerical Notes 2: Ordinary Differential Equations. Numerical Recipes Chapter on ODEs, first three sections
September 5 (Thursday): Finish ODEs, Start Root Finding and Minimization
- Readings: David Weinberg's Numerical Notes 3: Root Finding and Minimization. Numerical Recipes Chapter 9, focus on 9.0, 9.1, and 9.4
September 10 (Tuesday): No Class
September 12 (Thursday): No Class
September 17 (Tuesday): Finish Root Finding, Start Intro to Statustics
- Readings: David Weinberg's Statistical Notes 1: Background. Statistics, Data Mining, ... Chapter 3
September 19 (Thursday): Finish Intro to Statistics
September 24 (Tuesday): No Class
September 26 (Thursday): No Class
October 1 (Tuesday): Bayes Parameter Estimation
- Readings: David Weinberg's Statistical Notes 2: Bayesian Parameter Estimation
October 3 (Thursday): Finish Bayes, Start Stats 3
- Readings: David Weinberg's Statistical Notes 3: Errors, Likelihood, and MCMC
October 8 (Tuesday): No Class
October 10 (Thursday): No Class - Autumn Break
October 15 (Tuesday): Finish Stats 3
October 17 (Thursday): Stats 4 (Fisher Matrices)
- Presentation by Warfield: pandas
- Presentation by Pai: Gehrels (1986)
- Readings: David Weinberg's Statistical Notes 4: Fisher Matrices
- Readings: David Wittman's Fisher Matrix for Beginners
- Readings: Gould (2003) Chi^2 and Linear Fits
October 22 (Tuesday): No Class
October 24 (Thursday): No Class
October 29 (Tuesday): Stats 5: Hypothesis Testing
- Presentation by Duck: healpy
- Readings: David Weinberg's Statistical Notes 5: Hypothesis Testing
October 31 (Thursday): Finish Stats 5
- Presentation by Boley: scitkit-learn classification
- Presentation by Neustadt: astropy.table
November 5 (Tuesday): No Class
November 7 (Thursday): No Class
November 12 (Tuesday): Numerical 4: FFTs
- Presentation by Phillips: github
- Presentation by Duck: Isobe et al. (1990)
- Readings: David Weinberg's Numerical Notes 4: FFTs and Their Applications
November 14 (Thursday): Numerical 4: FFTs
- Presentation by Pai: scikit-lean dimensionality reduction
- Presentation by Warfield: Gott et al. (2001)
November 19 (Tuesday): Stats 6: Error Estimation
- Presentation by Neustadt: Busca and Balland (2018)
- Presentation by Phillips: VanderPlas (2017)
- Readings: David Weinberg's Statistical Notes 6: Estimating Errors From Data
November 21 (Thursday): Stats 6: Error Estimation
- Presentation by Boley: Forman-Mackey et al. (2012)
- Presentation by Phillips: VanderPlas (2017)
November 26 (Tuesday): No Class
November 28 (Thursday): No Class - Thanksgiving Holiday
December 3 (Tuesday): Finish Stats 6: Error Estimation - Last Class
- Presentation by Mayker: fits
- Presentation by Mayker: Kelly (2017)
Copyright © Paul Martini All Rights
Reserved