Bayequentist
-
Posts
37 -
Joined
-
Last visited
Posts posted by Bayequentist
-
-
The points mentioned by @BL4CKxP3NGU1N are all good. Additionally, you should also find out if your potential PhD university has a strong grad student union or not. Having a strong grad student union comes with many benefits like good health insurance, emergency fund, annual contract bargaining with the university etc... These "small" things make your life as a grad student (potentially 5+ years) much easier.
-
I have also just passed my PhD qualifying exam, and I think we share similar research interests. In the first year, besides the required statistics classes I also fit in my curriculum extra CS/EE classes like Information Theory, Convex Optimization, Machine Learning. As a consequence I was dead tired everyday, but I hope it will all be worth it? Learning from first principle is really awesome, but I definitely feel like it's not necessarily the optimal way to prepare for research. Thus I am also very interested to hear about more efficient ways to prepare for research!
-
Just now, DanielWarlock said:
It is not satire and it is a collaboration effort. Actually they may have eventually resolved to buy like 10 of these and put those in the prof's office for safekeeping because it is much cheaper if you buy a bunch at a time.
That sounds insane to me. How many people are there in your cohort that can pool together 2 million? Was there any funding or was it completely self-funded? Maybe a lot of people at Harvard are just independently rich? ?
-
I'd recommend one of the new ThinkPads with Ryzen 4000 CPUs that are coming out in the next 2 weeks. If you choose the L15 (starting at $649), you can stack up on RAM and SSD and the price would only be around $1k. The AMD's 7nm Ryzen processors, despite being cheaper, are thrashing Intel CPUs right now in terms of multi-core performance, which is very crucial if you do a lot of parallel computations. In terms of build quality and longevity, ThinkPads are the OG business laptops that will last you many years. GPUs are not really important, because even if you want to do deep learning, you'd want to use your university's computing cluster (my uni has a NVIDIA DGX-2 cluster available upon request) or use some cloud solution.
-
On a side note, does anyone know when will US News refresh the rankings for stat/biostat?
-
1 hour ago, GettingDemReactions said:
Their question is what COULD help their application. Both the Math GRE and contacting faculty COULD help them. Is it guaranteed? No, but you are spreading misinformation to say it can't help.
Agree that reaching out might help, but it won't help most of the time. From the pinned post by cyberwulf: Funding in most (but not all) U.S. stat/biostat programs is allocated at the department level to the strongest incoming students, so applicants aren't typically "matched" to potential advisors who agree to fund them*. Rather, the department projects the total number of positions available and then tries to recruit up to that number of students. Once the students are on campus, they are then either assigned to a position or (ideally) have some choices available to them. Of course OP should still try and reach out to faculty (but don't expect anything).
Regarding GRE subject test, OP did not take Abstract Algebra, Real and Complex Analysis, so taking the test will most likely mean throwing money away. Still, if OP is independently wealthy and willing to give it a shot then by all means go ahead and take the test.
-
Something to consider: given your strong background in math and stat (for a program like Oregon State), it should be no problem for you to get straight A’s in Casella & Berger.
-
In recent years I've seen quite a few stats PhD programs popping up that don't have coursework requirements for advanced statistical theory (Lehman & Casella...) and measure-theoretic probability (Durrett...) - e.g. programs that focus on Bayesian/computational/high-dimensional statistics and statistical learning. What are thoughts on those programs? Pros/cons in terms of academia/industry?
-
2 hours ago, bayessays said:
Should I delete this? It's an interesting change of pace.
Do you think OP is trolling or serious?
-
Admission committees in top schools receive a lot of >80% GRE scores from international students every year... so I'd say only submit it if they require it. 71% is pretty good for top 25~50 schools, but your profile is already competitive for those schools even without the score.
-
-
If you can show admission committee that you know linear algebra, I think you'd have a good chance of getting into Oregon State's MS Statistics program, with funding. It'd be a good stepping stone to a better PhD program.
-
I think DanielWarlock was a little confused. OP never stated that he wants a PhD in Statistics. Almost all of the materials taught in a typical MS Statistics program are not measure-theoretic (the probability course might touch upon a little bit of σ-algebra).
-
IIRC Stat PhD Now Postdoc mentioned that he did his PhD at UF.
You can search the database of gradcafe for GRE scores: https://www.thegradcafe.com/survey/index.php?q=uiuc+statistics and https://www.thegradcafe.com/survey/index.php?q=university+of+florida+statistics.
-
-
Aside from Mathematical Statistics, a course on Stochastic Processes is also a common follow-up to Stat 110.
Since galois seems to prefer watching lecture videos to reading books, fast.ai has a pretty interesting course on computational linear algebra.
-
52 minutes ago, Stat PhD Now Postdoc said:
These are very nascent fields (I've heard that even computer science conferences have difficulty finding suitable reviewers for papers on theory for deep learining) and so are fruitful areas for research. But since they are so recent, deep learning is not covered in The Elements of Statistical Learning (to the best of my knowledge -- there may have been a more recent edition of it than the one I read that contains sections on deep learning).
I would also note that ESL is by no means exhaustive. The authors are frequentist statisticians, so the book only very briefly touches upon Bayesian approaches to the same problems in the book (e.g. nonparametric regression, classification, etc.). But I think as a "gentle" introduction to a variety of machine learning methods and models, the book is quite good. IMO, this book is best utilized as a basic introduction and should be supplemented by reading other tutorials and lecture notes/slides and watching video tutorials.
The authors did publish a more recent book: Computer Age Statistical Inference, which has a better balance between frequentist and Bayesian approaches. There are also roughly 20 pages on Neural Networks and Deep Learning.
-
2 hours ago, Stat PhD Now Postdoc said:
"The Elements of Statistical Learning" is a good book and I found it to be pretty accessible. If you have already finished the first year of your PhD program, you probably can just start reading it (and then refer to other references when you need to). I think it's a good idea to read this book from start to finish at a high-level (i.e. read it enough to understand conceptually what's going on rather than focusing on the nitty gritty technical details). Then you can refer back to parts of it later and read those sections much more carefully if they are related to your research. I never did any of the problems at the end of the chapter, but I don't think that would have helped much for research anyway.
The best way to "start out" research is not to dive deeply into theory and technical details, but to get a broad sense of the field and then narrow the scope of your project (your PhD advisor should guide you to picking a suitable project that is not too ambitious and not too incremental but "just right" and doable enough to result in a paper or two). Reading through this book, watching tutorial videos, and reading review articles, lecture slides, tutorials, etc. are all great ways to learn the basics/high-level background of statistical learning. Once you start research, your advisor will probably give you a few important/"seminal" papers to read as well, and then it will make sense to delve into technical details, theory, etc.
Speaking of statistical learning, what do people in the stats community think of deep learning methods such as CNN, RNN, GAN, Deep Graphical Models? Thanks!
-
2 hours ago, salamiboyz said:
Should I give up then? Should I continue on taking the courses or taking upper level math classes? I am very set on this career path.
Not really. I am afraid and ashamed to show my GPA to people--it's a point of embarrassment for me so I seldom talk about it with advisors or my peers. I thought having my recommenders speak to my character and my ability to work hard would help make up for that in that aspect.
Could you list off some 30~50 biostats programs? What is the difference between these programs and, say, a program at Berkley or UCLA?These are the rankings that most people look at: https://www.usnews.com/best-graduate-schools/top-science-schools/statistics-rankings?name=biostatistics.
-
This is a tough one. Your low math grades are not consistent with your high quant GRE score. Do you have a professor that can explain the bad grades for you in his/her letter of recommendation? You are a URM and a female so you may still have a chance for a biostats program ranked 30~50, given your extensive research background, high GRE scores, and good LoRs.
-
2 hours ago, Logic said:
Ty for both of your responses. After giving some thought to things, I've decided not to apply for PhDs in computer science. Instead I'll focus on areas in physics and engineering (as that matches my background more appropriately) and pursue computational neuroscience research through those departments.
Your background is indeed enough for admission to Computational Science programs. You should head over to the Biology forum if you wanna ask about comp. neuro. programs! Most people on this forum only know about math/stats/biostats.
-
For Applied Math PhD, you'll need undergraduate Real Analysis.
For CS PhD, you'll need coursework on Data Structures and Algorithms (at least), and maybe Theory of Computation.
Though your mileage may vary, almost all competitive PhD applicants I know have the above-mentioned coursework in their transcript.
-
4 minutes ago, galois said:
I have typically learned math through textbooks but that stat110 looks like a very convenient format, especially given my time constraints, so I'll probably do that. Thanks for the advice.
For Schaums, I imagine since it's so popular, even if answers aren't in the back of the book, they're probably easy to find online?
Schaums has 2 sections: Solved Problems and Supplementary Problems. In Solved Problems, the solution is shown right after the problem statement. In Supplementary Problems, they only provide the answers to the problems without showing you how to solve them. I'd imagine you'd be able to find a lot of solutions to the Supplementary Problems online, but don't quote me on that:P
-
For Calculus: Schaum's Outline of Calculus is a classic book that has more than a thousand practice problems in it. The 5th edition is better than the 6th edition.
For Probability, Joe Blitzstein's class is awesome - I watched all the lectures and did most of the homework problems, so I can attest to the high quality of his course. If you still want a probability textbook, the standard recommendation would be A First Course in Probability by Ross. Though personally I've found Jaynes' Probability Theory: The Logic Of Science to be a much more enjoyable read. Jaynes is not strictly an introductory textbook, but if you've taken Real Analysis before then you should have no problem reading it, as he develops everything from the ground up.
UChicago vs CMU: Where would you go for a statistics PhD?
in Mathematics and Statistics
Posted · Edited by Bayequentist
I would take the results of the poll with a grain of salt because this question is heavily based on personal preference/ability.