Choosing MATLAB Toolboxes for Risk Analytics Work

Summary

The document considers which MATLAB toolboxes are useful for a risk analyst, including optimization, global optimization, econometrics, financial, and statistics tools, alongside database and market-data access. One practitioner reports that their team purchased only optimization and statistics toolboxes, preferring to implement additional functions themselves. The rationale is that many functions are manageable to build and doing so can deepen understanding and expose implementation pitfalls.

The response also describes a move from MATLAB to Python, citing its lack of licensing cost and the ease of sharing prototypes with front-office quantitative and technology teams. It points to common scientific Python libraries as building blocks and frames language choice as a systems integration decision. Another answer takes a contrasting view, saying nearly all of the listed toolboxes except econometrics are needed and recommending a risk forecasting reference. These are individual opinions, not a systematic comparison; appropriate tools depend on the analyst’s tasks, existing systems, and team practices.

Key ideas

• One risk analytics team reports buying optimization and statistics toolboxes while implementing other functionality internally.
• Writing analytical functions can help practitioners understand methods and identify implementation risks.
• Python is presented as a lower-cost, shareable alternative supported by scientific computing libraries.
• The responses disagree about how many MATLAB toolboxes are necessary, so requirements depend on context.

Tags

# What Matlab packages to I need as a Risk Analyst?


# What Matlab packages to I need as a Risk Analyst?












What toolbox are more suitable for a risk analyst. I found this:

- Optimization toolbox

- Global optimization toolbox

- Econometrics toolbox

- Financial toolbox

- Statistics toolbox

And also I have as a useful tool box:

- Database toolbox (I have some useful info in my database)

- Datafeed toolbox (I have a bloomberg)

Questions:

a. What of the first five toolboxes are actually useful for my purpose?

b. Is any other useful for my purpose? I have all approved but I don't want to abuse.

## Answer by Kiwiakos (score 2)

https://quant.stackexchange.com/a/16974

This is my experience (I am heading the Risk Analytics team for an investment bank in the City):

We only ever bought the Optimization and Statistics toolboxes. You are better off writing any extra functionality yourself. Most of the stuff is simple, and writing it yourself improves your understanding and highlights potential pitfalls.

Having said that, we recently abandonded Matlab and we have moved to Python. Not only free, but also makes it easy for us to share working code and prototypes with FO Quants and FO/Risk IT. But the points stand: using numpy, scipy and pandas you can build high quality libraries yourself.

This is something you might want to keep in mind. My understanding is that there is a general drive to integrate systems, and establishing a common language that IT understand is a strategic decision.

## Answer by Carlos S Jiménez (score 0)

https://quant.stackexchange.com/a/16172

Unfortunately you need all except econometrics toolbox. Look for a book called "Financial Risk Forecasting " by Jon Danielsson ,will be useful for what you need.

Shown in full with attribution under the source's licence. Licence: CC BY-SA 4.0 (Stack Exchange)

This summary was written by Stratmill's research agent from the original; it is not a copy of the source.