online calculus-based probablity & statistics course

[samiam]

Thanks so much. I will mention those words to the schools when inquiring about their courses.

I would go to NYC to start an MFE program but not to take a math course. It's sort of interesting. Some programs seem to say that a background with MV Calc and Linear Algebra combined with some programming know-how seems to cut the cake. However, that's not the sense I get after reading these boards.

Given my age (33) I feel it's sort of a little too late to start trying to collect more degrees to get that degree I really desire. I guess I will try to take in whatever courses I can handle, apply, and see what they say. I can offer the adcoms the conditional guarantee upon acceptance to take in other classes they want to see. However, I am really getting the picture now that this financial engineering is not so much about computer programming but really about high level math. As if the programming could be learned as you go along but the math you must pass a certain hurdle and that's at a minimum math major level at the bachelors but at a standard, a graduate degree in math.

And what's amazing on top of that is that the real quants are not the MFE's but the PhDs. An MFE is just like a quant grunt or quant-lite it seems. :-k

 

[bigbadwolf]

However, I am really getting the picture now that this financial engineering is not so much about computer programming but really about high level math. As if the programming could be learned as you go along but the math you must pass a certain hurdle and that's at a minimum math major level at the bachelors but at a standard, a graduate degree in math.

FE is a mix of scientific computing on the one hand and PDEs, stochastic, optimization, and econometrics on the other. The scientific computing (aka programming) is a crucial ingredient. A bachelor's with an emphasis in applied math should do the trick.

And what's amazing on top of that is that the real quants are not the MFE's but the PhDs. An MFE is just like a quant grunt or quant-lite it seems.

The right kind of Ph.D.s. Not a Ph.D. in pure math, but a Ph.D. in applied math, or physics, or even engineering, who can demonstrate experience with programming, numerical analysis, probability, and PDEs, and who can plug what holes may exist in his or her background (e.g., lack of stochastic or finance).

Also, it's an open question as to how relevant the high-level math is to understanding and predicting financial markets, and to pricing financial products. If you have the spare change, get hold of a copy of MacKenzie's An Engine, Not a Camera: How Financial Models Shape Markets.

 

[cstassen]

Working exclusively in the discrete setting, and the foundation for which is combinatorics. For example, I toss a coin seven times. What's the probability I get at least three heads? Or I have an urn with five green balls and seven red balls. I pick up two balls simultaneously. What are the odds both are red? This is the kind of material usually taught in European and Asian high schools and in the first year of US college courses (e.g., "College algebra and Probability"). If you have access to a university library, look for the book, "Elementary Probability with Applications," by Larry Rabinowitz (A K Peters, 2005); this covers non-calc probability. Of course, the moment one introduces a continuous setting, calc comes in.

Are you also asking what's the difference between calc-based probability and measure-theoretic probability?

Nono, I was just wondering bout the difference between calc based and non-calc based - what I meant with "talking about it" was what you just referred to, as it is more or less based on reasoning instead of methodology.
The difference between calc based and measure based was hammered into me last year:smt024

 

[jordanmvorsteg]

best calculus based probability class

I have two really good probability classes to choose from yet the first one is a thousand dollars less than the one that follows. Any advice?

STA 4321, 4322 - Introduction to Mathematical Statistics I and II (3-3)
This course presents an introduction to the mathematics underlying the concepts of statistical analysis. It is based on a solid grounding in probability theory, and requires a knowledge of single and multivariable calculus. Specific topics include the following: basic probability concepts, random variables, probability densities, expectations, moment generating functions, sampling distributions, decision theory, estimation, hypothesis testing (parametric and non-parametric), regression, analysis of variance, and design of experiments. Prerequisite: MAC 2313. (F,S)

or

STA 5446 - 5447 - Probability Theory I and II (3-3)
This course is designed to acquaint the student with the basic fundamentals of probability theory. It reviews the basic foundations of probability theory, covering such topics as discrete probability spaces, random walk, Markov Chains (transition matrix and ergodic properties), strong laws of probability, convergence theorems, and law of iterated logarithm. Prerequisite: MAC 2313.

 

[Hao]

5446-5447 are probability courses, and 4321-4322 are statistics. Usually, the statistical course starts from the BASIC cocepts of the probability. So, it really depends on what you need and what you want. BTW, 4321-4322 should be much easier than 5446-5447.

 

[cstassen]

Why don't you do all 4? They cover different topics! Will make you better! Second one sounds more applied (markovs etc), first one probably more difficult

 

[Winston]

Open CourseWare has the following course:

MIT OpenCourseWare | Electrical Engineering and Computer Science | 6.041 Probabilistic Systems Analysis and Applied Probability, Spring 2005 | Home

this looks like it has most of what's required