Probability and Markets [pdf] • Hacker News Board

Re: Probability and Markets [pdf]

#71

Post by KRAKRISMOTT » Sat, Jun 17, 2023, 6:14 AM UTC

lying4fun wrote: post #60

KRAKRISMOTT wrote: post #4

Earlier quoted context omitted.

Not Lie processes (are you referring to Lévy?), but stochastic calculus in general, which requires a strong intuition for differential equations, calculus, and statistics. Quantitative finance heavily favor those with a background in the physical sciences rather than discrete math like CS.

That is curious. Could you elaborate or point me to some sources, I would like to investigate that relationship further. I remember Jim Simmons saying that the Renaissance fund employs many physicist PhDs and other kinds of scientists. I thought they were there because of their specialisation but perhaps the reason is what you explained above or both

Physicists are popular because they deal with uncertainty and complex models across the time domain, which is another description of markets. When you zoom out far enough, a fluid dynamicist and a market making quant are not that different.

If you want books about trading mathematics and interviews in general, the green book (practical guide to quant finance interviews by Zhou) is the usual go-to. Falcon's Heard On the Street is another classic. Note both of these focus on generic interviews (the quant equivalent of leetcoding) usually for recruiting fresh grads directly out of school. You can probably skip some steps if you have the work experience or know the right people at certain funds.

If you want actual books on trading and modeling, that's another topic entirely.

Figure out which area specifically you are interested in: https://teddit.net/r/wallstreetbets/comments/kcs4xy/what_qua...

If you want an autobiography, Emanuel Derman's My Life as a Quant is well-regarded. I have never read it myself but Amazon has often recommended me The Physics of Wall Street.

Re: Probability and Markets [pdf]

#72

Post by kwant_kiddo » Sat, Jun 17, 2023, 7:18 AM UTC

KRAKRISMOTT wrote: post #4

kwant_kiddo wrote: post #2

During my university studies, I found it surprising that although some topics from advanced courses might be applicable in certain jobs like at Jane Street, interviews often emphasised the fundamentals and understanding of basic concepts such as basic probability theory, rather than some specialised knowledge in Lie processes. If you look through this PDF it does not even cover a 1.semester course in probability theo…

Not Lie processes (are you referring to Lévy?), but stochastic calculus in general, which requires a strong intuition for differential equations, calculus, and statistics. Quantitative finance heavily favor those with a background in the physical sciences rather than discrete math like CS.

yes I meant Lévy I don't really know what went through my head.

I also have to say that you don't need to have a strong intuition for differential equations even through they are very related. Stochastic calculus can be studied entirely in its own right.

And with respect to quantitative finance my experience is that while a candidate knowing about or having coursework in stochastic calculus is certainly relevant its not favourable.

This also depends on the exact job right, if you have to be pricing XVa products then you better have know Girsanov, Ito and what have you =)

Re: Probability and Markets [pdf]

#73

Post by smallnamespace » Sat, Jun 17, 2023, 7:27 AM UTC

kqr wrote: post #70

smallnamespace wrote: post #56

Earlier quoted context omitted.

> Ask people to predict whether the Jones' have a dog and they'll start asking questions about the specifics of the Jones', like do they have children, do they live in the suburbs, have they always wanted a dog, etc. Then they try to construct a narrative based on "logical" conclusions. This is the core idea of Bayesian statistics. You fall back to the background rate only if you have no ability to access more pertin…

When people ask about whether the Jones' have children, they do not (most of the time) have an accurate conditional probability of dog ownership given a certain number of children. They are trying to construct a story that seems plausible. The difference between Bayesian reasoning and narrative fallacy is a coherent evaluation of joint probabilities -- the bit most people skip over.

It's doubtful people have access to population-level base rates either, in which case they would also arrive at the wrong answer, no narrative needed.

But why do you point the finger at narrative instead of a simple lack of data? In the dog example, those questions are absolutely good ones to ask, so long as the data can be found.

One reason to reject this mental opposition between 'Bayesian reasoning' and 'narrative fallacy' is because narratives are essential to coming up with good conditions on which to split the population in the first place.

'I would ask whether they have children, because children like dogs and therefore it's plausible that families with children have a higher rate of dog ownership' is a reasonable narrative that justifies collecting an extra column in your dataset.

Granted, one shouldn't accept that story uncritically without checking against the actual statistics. But narratives themselves are unavoidable even when you're doing statistics correctly!

Re: Probability and Markets [pdf]

#74

Post by kqr » Sat, Jun 17, 2023, 8:48 AM UTC

smallnamespace wrote: post #73

kqr wrote: post #70

Earlier quoted context omitted.

When people ask about whether the Jones' have children, they do not (most of the time) have an accurate conditional probability of dog ownership given a certain number of children. They are trying to construct a story that seems plausible. The difference between Bayesian reasoning and narrative fallacy is a coherent evaluation of joint probabilities -- the bit most people skip over.

It's doubtful people have access to population-level base rates either, in which case they would also arrive at the wrong answer, no narrative needed. But why do you point the finger at narrative instead of a simple lack of data? In the dog example, those questions are absolutely good ones to ask, so long as the data can be found. One reason to reject this mental opposition between 'Bayesian reasoning' and 'narrative…

This is one of those things where it's impossible to prove one approach better or worse than the other. But history has shown (starting with merchant ship insurance in the 1700s going up to Tetlock's superforecaster research more recently) that starting from general base rates gives, on average, more realistic odds than starting from specifics.

You see this all the time in sports betting, too. People overcorrect from the base rate when receiving news. People love a narrative that makes sense to them, but statistically it rarely holds up against statistical reasoning with a very small set of variables. (For more examples, see Meehl's research around clinical vs actuarial judgment.)

Yes, you're correct that a hypothesis is one of those narratives, but if you're operating at that level of scientific abstraction you can ignore my comment. Most people don't, and just roll with the story and forget about the probabilities.

Re: Probability and Markets [pdf]

#75

Post by max_ » Sat, Jun 17, 2023, 6:40 PM UTC

Err_Eek wrote: post #19

antegamisou wrote: post #9

For the experts in the thread complaining about the simplicity of the excellent introductory text in the OP, here's an exciting reading for you: Statistical Consequences of Fat Tails: Real World Preasymptotics, Epistemology, and Applications by Nassim Nicholas Taleb https://www.amazon.com/Statistical-Consequences-Fat-Tails-Pr... Beware that it may be way more advanced than what you'd expect though.

Personally -- I couldn't get past the first 2 chapters of the book. The notations it introduces are pretty unfamiliar for a newcomer and it quickly becomes really hard to follow. I genuinely would like to hear from somebody who managed to go through the entire book above, and what their main takeaways were (from the chapters that follow the two introductory ones). The non-technical introduction chapter is pretty easy…

There is a Technical Incerto reading group (this book) that you may find useful.

https://www.techincertoreadingclub.com/