Looking for mentor for LOB study.

garpitg · Post by **garpitg** » Tue Nov 05, 2019 1:08 pm

Hi Folks,

I am U-HFT developer (Infra + Stratgies) having deep interest in unravelling information from LOB(Limit order book). I need someone who can push me deep into it.. (though I have started but feel lack of guidance). If anyone can spare his time to mentor me, that would be great. In return, I can do any technology-related work for you (I am an expert in C/C++/Linux/Python).

-Thanks

Arpit

trialanderror · Post by **trialanderror** » Tue Nov 05, 2019 1:54 pm

Hey,

Currently writing my MSc thesis on the subject. I'm following the paper I have attached, take a look. If you think it's too basic then never mind, otherwise I'm up for at least some discussion. Can't offer too much guidance though.

Regards,[url=/User%20Files/12832/LimitOrderBookInvestigation.pdf]Attached File: LimitOrderBookInvestigation.pdf[/url]

garpitg · Post by **garpitg** » Tue Nov 05, 2019 2:17 pm

Thanks for revert.. will read through it.. and get back in 2-3 days.

goldorak · Post by **goldorak** » Tue Nov 05, 2019 2:43 pm

I doubt a paper from 1995 being of any value in our century, especially when it comes to order books.

garpitg · Post by **garpitg** » Tue Nov 05, 2019 2:48 pm

I feel the fundamental properties should remain same. I am novice hence here, hence anything in direction is good for me..but Can you please provide your pointers, on how should I fullfill my learning thurst.

trialanderror · Post by **trialanderror** » Tue Nov 05, 2019 4:26 pm

@goldorak That's why I said it might be too basic, throw in old for good measure

JTDerp · Post by **JTDerp** » Wed Nov 06, 2019 11:27 pm

What's the end-game for you? Develop a systematic trading model for your resume/repo to show prospective employers, trade independently, intellectual curiosity, all of the above?

My (limited) 2c is that, if you really want to understand the dynamics, you need to broaden the view of analysis beyond the LOB, and include the (very situational) relations to the top-of-book returns time series, perhaps even intermarket and options order flow.

But then I know a guy who now works for IHT Markit and used to proclaim that you could predict the near-term direction of a futures market just from the stack/pull rates & distribution of the nearest 3 levels each side of the book...I was skeptical as to how robust that was.

Want some real fun, have a look at LOBs with Pro Rata matching (STIRs).

garpitg · Post by **garpitg** » Thu Nov 07, 2019 5:34 am

Hi JTDerp,

Thanks for revert.

Idea for this study is "trade independently & intellectual curiosity." I mostly work as independent consultant with prop firms to fund my research & data needs.

I understand when you say to look into options order flow, since the market I trade (Asian, Highly Volatile) options (European) cost structure is lesser hence order flow is trade more in that and it decides the course of action.

Stack/Pull rates is in my cards but distribution of levels is something I learned here.. I wil experiment with that and update my finding here.

Markets I focus on mostly FIFO, but it would be interesting to see details of Pro-rata.

HitmanH · Post by **HitmanH** » Thu Nov 07, 2019 10:44 am

Where are you based garpitg?

garpitg · Post by **garpitg** » Thu Nov 07, 2019 12:18 pm

Hi HitemanH,

I am in India. How about you ?