Best algotrading school

[Alexei_K]

Hello all,

I’d like to ask an advice which quant program I could choose.

A few words about me:
I have a decent experience in algorithmic trading and I wish to move forward in this industry, so the focus on algorithmic trading is important for me.
Another important feature is a good career service and a broad alumni network. The ideal school for me should have both qualities.

I will appreciate any comments and i will be grateful if you share your personal experience in PM.

 

[DMD]

The Baruch guys tend to know there stuff when it comes to practical skills and developing trading strategies...

 

[Byron]

Pick a big brand school at a good location (NYC or Chicago), you couldn't go wrong with that. Most quant funds (banks don't really do much algo trading nowadays) are strongly biased towards Ivy League or top 10 type schools.

 

[DominiConnor]

As far as I know, there isn't a dedicated algotrading masters course in the USA, there is one at UCL in London and that's pretty much it.
Jim Gatheral at Baruch has done some good stuff in this area, but their MFE is not optimised wrt HFT.

Be aware that Quant MFEs are mostly about things that have little to do with HFT and that assume things like no-arbitrage, memoryless time series, etc which an HFT trader assumes to not be true.

When I talk to people hiring HFT staff they talk about:
Unix internals
C
FPGA
Market Microstructure
Signal processing
Econometrics
Low level networking as well as FIX
Stats and data mining
Probability at a deep level
Game theory
AI
They generally go on a bit about wanting very good academic qualifications in general.

It's possible at some schools to do none of the incomplete list above, I don't believe there is an MFE which does even half of them.

Depending upon the background of the person I'm advising I often find that the best option may be a masters in Stats or Econometrics which are both easier to get into and cheaper.

 

[latency]

go to a brand name school, in the US-ivy league and top engineering schools, most hft firms recruit only from the top schools and a few schools around the chicago area. they dont expect you to know anything much, although having a computer science background puts you at an advantage.

 

[Siddharth Singh]

Will the Masters in Computer Science will fit more for the profile, as we can complete the additional requirements along with it. And I suppose it will keep a lot of options open.

 

[IlyaKEightSix]

The entire UWash comp finance master's program is built on the software written by the man who hired me in Chicago once upon a time, who's currently an algo trading partner at RCG securities, if that helps.

It's R-based though.

 

[albo]

As far as I know, there isn't a dedicated algotrading masters course in the USA, there is one at UCL in London and that's pretty much it.
Jim Gatheral at Baruch has done some good stuff in this area, but their MFE is not optimised wrt HFT.

Be aware that Quant MFEs are mostly about things that have little to do with HFT and that assume things like no-arbitrage, memoryless time series, etc which an HFT trader assumes to not be true.

When I talk to people hiring HFT staff they talk about:
Unix internals
C
FPGA
Market Microstructure
Signal processing
Econometrics
Low level networking as well as FIX
Stats and data mining
Probability at a deep level
Game theory
AI
They generally go on a bit about wanting very good academic qualifications in general.

It's possible at some schools to do none of the incomplete list above, I don't believe there is an MFE which does even half of them.

Depending upon the background of the person I'm advising I often find that the best option may be a masters in Stats or Econometrics which are both easier to get into and cheaper.

To summarize what you've written, I believe you're saying that one should be an expert in Computer Engineering, with a focus on Probabilistic Techniques applied to Economics. And you're also saying that this cannot be done by attending a regular Financial Engineering Program. I am interested to know which schools and programs might provide the highest percentage of the knowledge that you have described. Appreciate your opinion. Thank you.

 

[Alexei_K]

Hi Guys,

Thank you for all your responses, I really appreciate this. For the sake of the conversation let’s derive algotrading into quant trading and HFT. I agree about HFT preparation (Thanks for detailed response, especially for Domenic’s).
As a person with wide experience in developing quantitative trading strategies with large alpha (and less in HFT), I’d like to sharpen further my quantitative skills and fully utilize my potential. So the better focus on quantitative field the better. Namely, Statistical arbitrage (mean-reversion, momentum, regime-switching), Stochastic optimization, PCA/SVD, SVM.
I’ll appreciate any suggestions

 

[Anupriya]

To summarize what you've written, I believe you're saying that one should be an expert in Computer Engineering, with a focus on Probabilistic Techniques applied to Economics. And you're also saying that this cannot be done by attending a regular Financial Engineering Program. I am interested to know which schools and programs might provide the highest percentage of the knowledge that you have described. Appreciate your opinion. Thank you.

Hi,

I would like to inform you about a 4-month long comprehensive programme in Algorithmic Trading offered by QuantInsti (QI), an Institute in Quantitative Trading, based out of Mumbai, India. This course is also offered online, aimed at working professionals with 3-hrs sessions on Saturday-Sunday and recordings are available for later viewing.

The course can be broken down primarily into three components:
(i) Advanced statistics & high-frequency econometrics
(ii) Quantitative Trading Methodologies (specific focus for strategies for different asset classes)
(iii) Software, architecture & financial computing (including programming) for HFT trading.

The detailed list of topics is uploaded for your reference.

Apart from imparting knowledge about advanced concepts within high-frequency trading, QI also attempts to provide practical insights into aspects like system architecture & latency, standardized protocols, trading strategy design methodologies for HFT, risk management for HFT, and new developments / tools in this domain.

QuantInsti is a venture by iRageCapital Advisory Private Limited. iRageCapital provides the complete spectrum of services associated with Algorithmic Trading (both Trading Technology & Quant Strategy Development) to both sell-side & buy-side institutions.

Thanks and regards,
Anupriya
Ph: +91 986 785 0080
www.quantinsti.com

 

[darth]

I though Andy had a different section for advertisements...

 

[AndrewChang]

Hi Guys,

Thank you for all your responses, I really appreciate this. For the sake of the conversation let’s derive algotrading into quant trading and HFT. I agree about HFT preparation (Thanks for detailed response, especially for Domenic’s).
As a person with wide experience in developing quantitative trading strategies with large alpha (and less in HFT), I’d like to sharpen further my quantitative skills and fully utilize my potential. So the better focus on quantitative field the better. Namely, Statistical arbitrage (mean-reversion, momentum, regime-switching), Stochastic optimization, PCA/SVD, SVM.
I’ll appreciate any suggestions

I think that you will find the Baruch program extremely well suited for your taste. Domini mentioned that he doesn't believe any program even provide half of that list but I am here to tell you that Baruch really teaches you 6/11 ( more than half ) of that list . One of the great values of the program is to learn market microstructure from someone who is extremely active in that field (not to mention the fact that I don't think any any other program offers it other than Steven's institute of technology). We have a class statistical arbitrage class where you implement trading strategies and just compete against each other where the winners present to PDT and SAC (it was pretty awesome, 2 of the most amazing firms). We do stochastic optimization in market microstructure for execution strategies, SVM in machine learning, and PCA practically in every single class. But what I think the best thing about Baruch's program is the career service (I really believe it is second to none because the director really provides great career service ). I believe we have a great track-record for students that want to go into trading as everybody that wanted to go into trading got a job in it (I intern at State Street working on quant strategies in energy futures and will be working full time at as an options market maker next semester and many of my classmates have way cooler jobs).

 

[dstefan]

Andrew - one correction: the program director does not get students jobs. We provide the best education and interviewing opportunities to our students, but it them who get the jobs.

 

[TheDemonKing]

As far as I know, there isn't a dedicated algotrading masters course in the USA, there is one at UCL in London and that's pretty much it.
Jim Gatheral at Baruch has done some good stuff in this area, but their MFE is not optimised wrt HFT.

Be aware that Quant MFEs are mostly about things that have little to do with HFT and that assume things like no-arbitrage, memoryless time series, etc which an HFT trader assumes to not be true.

When I talk to people hiring HFT staff they talk about:
Unix internals
C
FPGA
Market Microstructure
Signal processing
Econometrics
Low level networking as well as FIX
Stats and data mining
Probability at a deep level
Game theory
AI
They generally go on a bit about wanting very good academic qualifications in general.

It's possible at some schools to do none of the incomplete list above, I don't believe there is an MFE which does even half of them.

Depending upon the background of the person I'm advising I often find that the best option may be a masters in Stats or Econometrics which are both easier to get into and cheaper.

Hi DominiConnor,
I'm currently working as an actuarial intern and I find it boring. If I were to go to the University of Maryland Applied Math program and learn almost all of those classes you mentioned above along with a few financial classes and math classes associated with finance - stochastic processes, partial differential equations, real analysis, etc, do you think I would be able to find a job as an algorithmic trader? Do you think I might be better off getting a PHD and going deeper into one of those fields you specified above? Or should I just go to a math finance program and try to take some extra computing classes?

 

[AndrewChang]

Andrew - one correction: the program director does not get students jobs. We provide the best education and interviewing opportunities to our students, but it them who get the jobs.

I stand corrected

 

[arsalan khan]

As far as I know, there isn't a dedicated algotrading masters course in the USA, there is one at UCL in London and that's pretty much it.
Jim Gatheral at Baruch has done some good stuff in this area, but their MFE is not optimised wrt HFT.

Be aware that Quant MFEs are mostly about things that have little to do with HFT and that assume things like no-arbitrage, memoryless time series, etc which an HFT trader assumes to not be true.

When I talk to people hiring HFT staff they talk about:
Unix internals
C
FPGA
Market Microstructure
Signal processing
Econometrics
Low level networking as well as FIX
Stats and data mining
Probability at a deep level
Game theory
AI
They generally go on a bit about wanting very good academic qualifications in general.

It's possible at some schools to do none of the incomplete list above, I don't believe there is an MFE which does even half of them.

Depending upon the background of the person I'm advising I often find that the best option may be a masters in Stats or Econometrics which are both easier to get into and cheaper.

This is far by the best advice i got. I thought Quant had everything to do with HFT. i assume what you are saying is getting a masters in Econometrics could land you a job in HFT?

 

[Alexei_K]

Hi Andrew,

Thank you for very insightful answer, it was very interesting for me to learn more about your personal Baruch experience. And with regard to mentioned PCA and SVM vehicles, their business interpretation should be very exciting, especially in Market-micro structure context, lead-lag strategy comes in mind.

Also you mentioned about statistical arbitrage class;

• What facilities have you had in terms of hardware and available market data?
• How is the PDT and SAC competition look like?

Alexei

 

[Licia]

Hi, if you want a well rounded crash-course you can look at our QTrade bootcamp programme www.qtradebootcamp.com
There is a nice balance of theory, coding and real-market strategies.
BR,
Licia