HELP! PhD / MS course preparation - advice/comments

[hedgie]

I am planning out my next two years of Math/Stats course work before applying to a Statistics PhD program. My background....I have done two undergrad degree's the first one was miserable and was years ago...I assume most adcom's will look past that.

By the end of this summer I will have completed:

Cal I-III

Intro Linear Algebra

Discrete

Probability Theory

Applied Differential Equations I

Programming I (Python)

Econometrics **(Don't know that this is relevant)

Econ Forecasting **(Don't know that this is relevant)

Here is what I am looking take/do (the courses in italics I figured were what I should take:

This is for the coming summer, undecided here but believe the intro to scientific computing is more relevant:

Intro to Scientific computing (believe this is Mathematica based course based on numerical analysis) or Programming II (Python)

Fall 2011

Intro to Real Analysis I

Unsure what to take in the fall semester here are the options:

Math Stats w/applications I **(More rigorous option in Masters program but this is required for stochastic processes in the spring, I don't know if taken math stats twice is good or bad??)

Numerical Analysis I

Linear Optimization Theory (linear programming)

Applied Differential Equations II (recommend by grad advisor for applied math degree, but maybe not applicable for Stats PhD?)

Foundations of Computer Science

Programming III C++ (If I took Programming II in the summer)

Spring 2012/Summer 2012

Advanced Linear Algebra

Intro to Real Analysis II

Advanced Calculus

Principles of Modern Algebra I (recommended by grad advisor)

Stochastic Processes w/Applications I

Math Stats w/Applications II

Data Structures and Algorithms for CS

**Are four math courses a recipe for disaster? As well I have conflicting advice from advisors, professors, and PhD students on the Modern Algebra. It seems adcoms want heavier math but the math is rarely used....as I mentioned I have a strike against me due to my first poor showing in undergrad years ago.

Fall 2012-Summer 2013 (MA in Applied Math)

Real Analysis I & 2 (definitely taking)

Numerical Linear Algebra

Intro to Complex Calculus (this can be taken as an undergrad too, one of two courses offered in summer; only can take it as an undergrad if people think it is advisable to take 4 courses in one of the semesters)

Stochastic Processes w/Applications II (only if I could handle 4 courses in a semester)

Graph Theory with Applications (It's one of two courses offered in the summer, gets me out in a year)

I am required to take two of these sequences:

Math Stats I (more advanced than the undergrad version) (undergrad or grad?)

Numerical Analysis I & 2

Linear Optimization I & Non Linear Optimization Theory I

Is it feasible/logical to take four math courses a semester in grad school? I assume one could do it and get the grades....but I want to learn the material. If not that rules out Stochastic applications, and Numerical Linear Algebra or RA 2.

Opinions, advice, thoughts, are all appreciated! Thanks for taking the time to look at my post. As well I could do a MS in stats instead of Math, but I assumed Math Stats I & II, Numerical LA, RA 1 & 2, and stochastic processes were almost two thirds of the degree and would help strengthen my application. I am trying to apply to a more theory oriented program and the Master's here would be applied and less math intensive i.e. no RA required. I have been advised by various adcom's to take math and I will learn stats when I get there...other PhD's have advised the opposite...learn linear algebra, RA, and stats.

Thanks in advance, I know it is a lot but I have spent considerable time trying to come up with some sort of optimal solution to get in a good program. I currently have a math GPA of 3.75, with one B. If successful and assuming I can maintain a 3.75 or better and achieve a 3.5 or better in grad school what tier programs could I look at applying to. I assume I can bring the math GPA up as it had been some years since my last math course (10ish years). I am set on nothing and would scrap and start over...just trying to give my best shot at getting in a good program.

Thanks again for taking the time out of your busy schedule to read this!

Cheers!

[Matt W]

I would definitely recommend taking real analysis. This is essential since you're planning to apply to a more theoretical program. Likewise, linear algebra is also important.

It's also a good idea to take stats courses for your electives.

Other than that, none of the other courses seem essential. However, doing well in advanced math courses will strengthen your application. You might want to take courses which will let pursue a PhD in either pure or applied math in case you change your mind - you still have a ways to go before you graduate.

It is fairly common for people without assistantships to take 4 grad courses per semester.

[hedgie]

Thanks! I had planned on definitely taking Intro to Real Analysis I & II and Real Analysis I & II in the master's program.

I had also definitely planned on taking the Advanced Linear Algebra, and the Numerical Linear Algebra. In regards to the Math Stats I & II do you have any advice/opinion regarding my dilemma. I can take Math Stats I & II w/applications as an undergrad course....then/or take Math Stats I & II as a grad course (two different series courses) which is more theory and requires 'Advanced Calculus' and by 'Advanced Calculus' that is not a Real Analysis course at my school...more of an engineering applications based Advanced Calculus beyond multi variable, Cal III.

Would you think to take both sequences, or just the graduate? I figure I didn't want to waste 6 credit hours repeating coursework. I had also heard that Stochastic Processes was worthwhile and the undergrad Math Stats would be required for me take both sequences of it, if not I could complete one of the two out of the Stochastic sequence.

Thanks I greatly appreciate the advice/thoughts! I had kind of thought of leaving open the PhD in Math route even though I am pretty set on Stats, but that was the reason I listed the Advanced DE 2, Modern Algebra, Complex Analysis, etc....

Thanks again!

Edited February 6, 2011 by hedgie

[Matt W]

Math Stats is a fairly general term, so it's hard to give definite advice. If you will be covering much of the same material, then there is little point in taking both sequences. It would be best to compare course descriptions or to ask a professor to find out how much overlap there is.