How to get a non-quant up to speed?

[Ken Abbott]

Ok, forum. This time I need your advice.

A colleague of mine is on the management fast track. She's smart, personable, and talented. Her challenge is that she has no formal quant training at all. She is destined to be part of senior management. A minimal amount of schooling would help her career tremendously.

My problem here is that I don't have a good window into "basic training" any more. In my day (the 80's and 90's) you took calc 1, calc 2, inference, probability, stochastic processes, bond pricing. The game, however, has clearly changed. Between TED talks, Kahn Academy, and the whole MOOC thing, there are myriad ways for someone to get the basics without enrolling in a formal classroom program.

What would you recommend?

 

[bigbadwolf]

Ok, forum. This time I need your advice.

A colleague of mine is on the management fast track. She's smart, personable, and talented. Her challenge is that she has no formal quant training at all. She is destined to be part of senior management. A minimal amount of schooling would help her career tremendously.

How minimal? Specify.

My problem here is that I don't have a good window into "basic training" any more. In my day (the 80's and 90's) you took calc 1, calc 2, inference, probability, stochastic processes, bond pricing. The game, however, has clearly changed. Between TED talks, Kahn Academy, and the whole MOOC thing, there are myriad ways for someone to get the basics without enrolling in a formal classroom program.

What would you recommend?

A personal mentor with genuine insight. And some insightful books. No matter how you specify "minimal," it's going to take time, effort, good books, and at least one good teacher. Kahn, MOOCs, TED won't do the trick -- unless you already know what you're looking for. Even then, the lack of that alchemical ingredient, human contact, bothers me.

 

[Daniel Duffy]

Quants spend > 60% of their time programming AFAIK.

As senior manager maybe this not needed.

 

[bigbadwolf]

Quants spend > 60% of their time programming AFAIK.

As senior manager maybe this not needed.

But today those senior managers need to understand the world of coding -- they need to have done coding at some point in their careers.

 

[Ken Abbott]

Quants spend > 60% of their time programming AFAIK.

As senior manager maybe this not needed.

Most of the quants I know spend little time programming. Those are the ones in the support functions- risk management, model review, price verification, audit.

 

[Barny]

How much maths has she done?

If she hasn't done even basic calculus or algebra then I fail to see how she can understand anything in a time-frame of less than 5 years.

 

[Daniel Duffy]

Most of the quants I know spend little time programming. Those are the ones in the support functions- risk management, model review, price verification, audit.

Sounds like you have low software maintenance costs. What happens if you need a new model?

 

[Ken Abbott]

Sounds like you have low software maintenance costs. What happens if you need a new model?

We get a model developer to develop one. I wouldn't let most quants anywhere near production code, unless I only needed the price once. If I need that price 10,000 times x 1,000,000 positions, I would give it to a programmer who could price efficiently.

 

[soufiane]

I can suggest her this course : Financial Engineering and Risk Management taught by Martin Haugh and Garud Iyengar of Columbia University
https://class.coursera.org/fe1-001/wiki/view?page=syllabus
Monte Carlo Methods in Finance :
https://moocfellowship.org/submissions/monte-carlo-methods-in-finance

 

[Daniel Duffy]

We get a model developer to develop one. I wouldn't let most quants anywhere near production code, unless I only needed the price once. If I need that price 10,000 times x 1,000,000 positions, I would give it to a programmer who could price efficiently.

My remark is a bit OT, but a good developer who understands the math+finance as well is a great skill to have.

 

[bordingj]

I have exploited alot of the online free education stuff with great success. I use MOOC etc when I want to learn more than what my own University provides, when I have had to review some material or when some material is better explained at courses on the web than in a book or by some of my professors.
Depending on her current level here is my recommendation. All courses listed are completely free.

For basic algebra, calculus, linear algebra, probability and statistics:

• Watch Khan Academy videos and do pen and paper exercises & problems (get some good books) (Salman Khan is without doubt one of best instructors in the world.)
• Take "Calculus 1" from Ohio State university at coursera https://www.coursera.org/course/calc1
• Do the Linear algebra course from Texas at Austin at edx https://www.edx.org/course/utaustin/ut-5-01x/linear-algebra-foundations/1162
• Statistics courses stat2.1 and stat.2.2 and stat2.3 from Berkeley at edx
  • https://www.edx.org/course/uc-berkeley/stat2-1x/introduction-statistics/594,
  • https://www.edx.org/course/uc-berkeley/stat2-3x/introduction-statistics/825,
  • https://www.edx.org/course/uc-berkeley/stat2-2x/introduction-statistics/685

For basic programming (basic programming is recommended/needed for doing exercises in basic quant finance courses):

• Do "Learn to Program: The fundamentals" offered by University of Toronto at coursera https://www.coursera.org/course/programming1 or
• "Introduction to Computer Science Course" from MIT at edx https://www.edx.org/course/mit/6-00-1x/introduction-computer-science/1122
  
• Then do "Algorithms, Part 1" course offered by Princeton at coursera https://www.coursera.org/course/algs4partI

For more advanced calculus and basic differential equations:

• Do "Calculus 2: Sequences and Series" offered by Ohio at coursera https://www.coursera.org/course/sequence
• Multivariate calculus (Calculus 3) lectures are available at https://www.udemy.com/calculus-iii-multivariable-calculus/ and at MIT OCW http://ocw.mit.edu/courses/mathematics/18-02sc-multivariable-calculus-fall-2010/
• "Differential Equations" course at MIT OCW http://ocw.mit.edu/courses/mathematics/18-03sc-differential-equations-fall-2011/
• Then do the "Mathematical Methods for Quantitative finance" course from Washington university at coursera.org https://www.coursera.org/course/compdata (they use the book "A primer for the mathematics of financial engineering" by Stefanica)
• Watch Khan Academy videos on the subjects if needed (and available)

For basic numerical methods

• A good read in basic numerical methods (for example "Applied Numerical Methods with Matlab for Engineers and Scientists" by Chapra)
• "Introduction to numerical analysis" from MIT OCW http://ocw.mit.edu/courses/mathemat...n-to-numerical-analysis-spring-2012/Syllabus/ (no video lectures available - only notes and assignments)
• Then do the course "Scientific Computing" by University of Washington at coursera https://www.coursera.org/course/scientificcomp

Linear programming and Optimization

• Do "Linear and Integer programming" course at coursera https://www.coursera.org/course/linearprogramming

For fundamental stochastics (I haven't found any good online source for this)

• Study "Measure, Integral and Probability" by Capinski and Kopp
• Then study "Basic Stochastic Processes" by Brzezniak and Zastawniak

For basic quant finance:

• Do the courses "Financial Engineering and Risk Management" Part One and Two from Columbia University at coursera
  • https://www.coursera.org/course/fe1
    
  • https://www.coursera.org/course/fe2
• Do the course "Introduction to Computational Finance and Financial Econometrics" from Washington at coursera https://www.coursera.org/course/compfinance
• Do "Monte Carlo Methods" course at iversity https://iversity.org/courses/monte-carlo-methods-in-finance
• Study "Investment Science" by Luenberger
• Study "Paul Wilmott introduces Quantitative Finance"
• Study Shreve's "Stochastic Calculus for Finance I: The Binomial Asset Pricing Model"

More quant finance

• Study "Stochastic Calculus for Finance II: Continuous-Time Models" by Shreve
• Do the "Asset Pricing" course from University of Chicago at coursera https://www.coursera.org/course/assetpricing (this is an advanced course!)

Getting a good grasp of everything above would take at least 2 years of full-time study if you are smart, I think (given that some of it will be a review).

 

[Ken Abbott]

I have exploited alot of the online free education stuff with great success. I use MOOC etc when I want to learn more than what my own University provides, when I have had to review some material or when some material is better explained at courses on the web than in a book or by some of my professors.
Depending on her current level here is my recommendation. All courses listed are completely free.

For basic algebra, calculus, linear algebra, differential equations, probability and statistics:

• Watch Khan Academy videos and do pen and paper exercises & problems (get some good books) (Salman Khan is without doubt one of best instructors in the world.)
• Take "Calculus 1" from Ohio State university at coursera https://www.coursera.org/course/calc1
• Do the Linear algebra course from Texas at Austin at edx https://www.edx.org/course/utaustin/ut-5-01x/linear-algebra-foundations/1162
• Statistics courses stat2.1 and stat.2.2 and stat2.3 from Berkeley at edx
  • https://www.edx.org/course/uc-berkeley/stat2-1x/introduction-statistics/594,
  • https://www.edx.org/course/uc-berkeley/stat2-3x/introduction-statistics/825,
  • https://www.edx.org/course/uc-berkeley/stat2-2x/introduction-statistics/685

For basic programming:

• Do "Learn to Program: The fundamentals" offered by University of Toronto at coursera https://www.coursera.org/course/programming1
• Then do "Algorithms, Part 1" course offered by Princeton at coursera https://www.coursera.org/course/algs4partI

For more advanced calculus etc:

• Do "Calculus 2: Sequences and Series" offered by Ohio at coursera https://www.coursera.org/course/sequence
• Do the Mathematical Methods for Quantitative finance course from Washington university at coursera.org https://www.coursera.org/course/compdata (they use the book "A primer for the mathematics of financial engineering")
• Watch Khan Academy videos on the subjects if needed (and available)

For numerical methods

• First do "Differential Equations in Action" at Udacity https://www.udacity.com/course/cs222
• Then do the course "Scientific Computing" by University of Washington at coursera https://www.coursera.org/course/scientificcomp

Linear programming and Optimization

• Do "Linear and Integer programming" course at coursera https://class.coursera.org/linearprogramming-001/class

For fundamental stochastics (I haven't found any good online source for this)

• Study "Measure, Integral and Probability" by Capinski and Kopp
• Study Shreve's "Stochastic Calculus for Finance I: The Binomial Asset Pricing Model"
  
• Then study "Basic Stochastic Processes" by Brzezniak and Zastawniak

For basic quant finance:

• Do the courses "Financial Engineering and Risk Management" Part One and Two from Columbia University at coursera
  • https://www.coursera.org/course/fe1
    
  • https://www.coursera.org/course/fe2
• Do the course "Introduction to Computational Finance and Financial Econometrics" from Washington at coursera https://www.coursera.org/course/compfinance
• Study "Investment Science" by Luenberger

More quant finance

• Do the course Monte Carlo Methods in Finance at iversity https://iversity.org/courses/monte-carlo-methods-in-finance
• Study "Stochastic Calculus for Finance II: Continuous-Time Models" by Shreve
• Do the "Asset Pricing" course from University of Chicago at coursera https://www.coursera.org/course/assetpricing (this is an advanced course!)

Getting a good grasp of everything above would take about 2 years of full-time study if you are smart i think (given that some of it will be a review).

Thank you. This is perfect.

 

[systemtrader]

There's also MIT OCW:
http://ocw.mit.edu/index.htm

Single Variable Calculus:
http://ocw.mit.edu/courses/mathematics/18-01sc-single-variable-calculus-fall-2010/index.htm

Multivariable Calculus:
http://ocw.mit.edu/courses/mathematics/18-02sc-multivariable-calculus-fall-2010/index.htm

Linear Algebra:
http://ocw.mit.edu/courses/mathematics/18-06sc-linear-algebra-fall-2011/index.htm

Differential Equations:
[url]http://ocw.mit.edu/courses/mathematics/18-03sc-differential-equations-fall-2011/index.htm[/URL]

 

[Daniel Duffy]

Quote: For fundamental stochastics (I haven't found any good online source for this)
Stochastics in still in its infancy, especially when it gets to computation (the numerical methods are not robust, e.g. Euler, Milstein).

IMO a good aproach will be finance+maths+numerics+applications+programming so that understand A-Z.

Stochastics presupposes some mathematical sophistication in measure theory, Lebesgue theory, probability spaces and some Functional Analysis (e.g. normed linear spaces and inequalities). Without this basis it just looks like a bunch of symbols that you have to get used to.

And I reckon RNG, numerical solution of ODEs as well.

This could be the reason for this phenomenon.

When you can do/understand the maths _and_ can program it the you understand it.

http://www.sitmo.com/article/popular-stochastic-processes-in-finance/

 

[jeanpeyre]

I realize this is an old post but I can recommend this new stochastic calculus course in udemy

udemy stochastic calculus

 

[ExSan]

Quote: For fundamental stochastics (I haven't found any good online source for this)
Stochastics in still in its infancy, especially when it gets to computation (the numerical methods are not robust, e.g. Euler, Milstein).


When you can do/understand the maths _and_ can program it the you understand it.

http://www.sitmo.com/article/popular-stochastic-processes-in-finance/

 

[jeanpeyre]

In fact there are 2 courses in udemy covering stochastic calculus and option pricing

Udemy Stochastic Calculus
Udemy Math Option Pricing

The stochastic calculus covers martingales brownian motions girsanov theorem...
The math option pricing covers the pricing of vanilla options, some european payoffs, local volatility...