takemoremath

Not that I necessarily think this is a good idea but actually the math on the GRE subject test is not so advanced. According to ETS, 50% is from single and multivariate calculus and 25% from linear algebra. It is only the remaining 25% of the content that is drawn from more advanced topics. So someone who remembers their calculus quite well could get a decent score.

Detexify is amazing for locating symbols: Detexify LaTeX handwritten symbol recognition I also really like the LaTeX Wikibook: LaTeX - Wikibooks, collection of open-content textbooks It is quite comprehensive and a lot easier to navigate than the more classic references.

I'm not sure if it will help you at all for applying. But it is possible to do well on the test without lots of advanced math. If you read the fine print about the exam most of the material is based on freshman/sophomore level math. 50% of the exam is based on single and multivariate calculus, 25% on linear algebra, and the remainder on "advanced" topics such as analysis, differential equations, topology, etc. Though the mathematical content is not particularly advanced be warned that the questions are often quite tricky.

One advantage of a stat degree vs. a pure math degree is that a stat masters has far better private sector employment opportunities. So if along the way your plans change a stat degree is good credential to have. Plus, it is also my impression that funding comes easier for MS stat than MS math.

If I understand what you are doing correctly, I think this result is best achieved with the enumerate environment rather than \section and \subsection commands. There are many examples online. Just google "Latex enumerate environment." To get sub-problems you just nest one enumeration inside of another.

The LaTeX wikibook is also a very good LaTeX reference. It's pretty well developed, easy to navigate, and covers everything you need to know. I find it much easier to use than the sprawling PDF manuals available elsewhere online.

A nice feature of analysis, if viewed in the right way, is that it is a very intuitive field of mathematics. The best analysts I know do have a sort of "talent". But to me this talent is really just a deep conceptual understanding of the material. They let their intuition guide them through tough proofs, looking at the big picture to try to connect the pieces. Such intuition, can be developed by first obtaining a solid understanding of the definitions and concepts and then by organizing the ideas into some kind of conceptual framework. In this way, working through the text a bit on one's own can serve as good preparation. Of course, the student should be paying attention to things they don't understand at first pass, perhaps writing them down and making sure they are adequately addressed during class. And certainly this preparation must include attempts at the proofs in the text and the exercises. But this is a mathematicians perspective...

Why not find out what text the class uses and read ahead? Try working as far through the book as you can on your own. Get a feel for both the language and technique of analysis. Make notes of concepts and proofs you don't understand. Hopefully, these will be illuminated during the term. Good luck.

Have you considered taking a year or two off? I ask because you're really throwing a lot of ideas around about what you want to study and why. Some time away from grad school might help you develop your focus...and probably give you a much better sounding back story. And of course the worst thing that could happen to you is that you jump from one poor fitting program into another...

At many schools Calculus IV is a first course in ordinary differential equations.

I think you might also see difference equations in a course on numerical methods for differential equations. This would be a very specialized course though. I would suspect that if you have a good background in differential equations, you can pick up all you need to know about difference equations.

Ok. But what if you thought you would be in the top of the class at Albany and only in the middle (or lower) at UVA? Should this make a difference in your choice?

Thanks for the suggestions...but I ended up answering my own question. I found the Dover books in ODE and PDE, respectively, are sufficient for my purposes.

Does anyone have a favorite Diff-Eq text? I'm looking for something which is a nice reference for methods in ODE and PDE.

You might want to spend some time thinking about what types of jobs you are interested in at these places (and others). An MS in statistics is currently a very flexible and marketable degree which would leave you with many employment options. But you're not likely to get specific training in financial or public policy analysis on this route, which may be important for some positions at the places you named. On the other hand, a degree in applied econ or public policy may be more limiting in terms of opportunities outside of economic and policy analysis. The more specific you can get about what you want to do, the easier it will be to decide which degree program is better.

It sounds like you would also be well prepared for a masters in statistics. There are several reasons to consider this. First, it's likely to be a better fit than a masters in math (you probably have all the pure math you'll ever need, but probably not all the statistics training). Second, your funding opportunities may be considerably better in stats.

I don't know about this. I think many people know that the actuarial exams are very competitive (25% pass rate?) and that passing them is significant. I think it's probably a detail to put on your CV/Resume but not make too big a deal out of otherwise.

I've heard about people typesetting during class, but I have never seen it done. I would guess that the difficulty of typesetting during class depends on the material itself. If you are typesetting stuff that is mostly words with equations and symbols interspersed here and there, Latex would be very efficient in class. And it should be very easy to do this. You can always type without compiling and then debug minor errors after class. But typesetting long equations and arrays is very tedious and you're likely to make many errors if you rush. Particularly if you have lots of equations involving nested arguments or need to type matrices (ugh!) it seems like it would just be too difficult.

I would say it's no harder than using SAS or MatLab. It just takes some getting used to and it's much less "point and click" then Stata or eviews. If you've done any type of programming (C++, Java) it shouldn't be a challenge. Regardless of the package, the hardest part is always doing the statistics right!

I used TSP several years ago. Actually, I think it is one of the first econometric software platforms that was available on PC's. In the 90's TSP split up into TSP International (which publishes the current TSP versions) and the Eviews company. TSP is a language (based around C++) and you write programs to run your routines. If you are good at those types of things you might find it fairly easy to pick up; once you learn the basics. Generally, I'd recommend using Stata over TSP. Stata is far easier to learn and use, and I think it is basically the standard package for econometrics. Anything you'd want to do in TSP can certainly be done more easily in Stata...and Stata is far more powerful and adaptable. But maybe you have some compelling reason for using TSP over other programs?