Recent content by bluelou

I actually think it's a bit better, or maybe I'm just participating in a smarter way. I've had a few substantive chats recently, where people actually learned something and there were few spammers. Is this the best site for quants or even institutional traders? No, but there's nothing wrong...

I can't speak for US equities, but for CME IQ Feed has worked for me for years for L1. Great match to institutional benchmarks. No, I wouldn't rely on it for really low latency stuff (e.g. colocation), but it's good enough for an active trader feed.

My choice of words wasn't helping. Sorry, I've made this more confusing than it needs to be. Let's start with an image. Looking at the previous plot labeled 'risk-averse', think of the bottom right part of the utility curve as anchored at the origin and the upper right part being bent...

"Increasing the risk aversion parameter (smaller Kelly fraction) would reduce the curvature of the risk-averse utility curve." My apologies, this sentence is incorrect. This would _increase_ the curvature, not reduce. That is, it would increase the concavity.

Have you benchmarked the data? What did you find? Thx.

You're welcome. Regarding your examples, think of utility (e.g. Kelly) as _your_ risk preference. The examples you've described are _your_ risk preferences for the 3 cases. A typical casino game fits with risk-affine utility. You already know the house has an edge but that's not stopping you...

Has algoseek improved? I checked it out a few years ago and thought the CME Level 1 data was pretty low quality. IIRC, > 70% of the quote messages were missing vs Reuters and some other benchmarks.

The math and assumptions are critical when working with utility functions like the Kelly Criterion, but I think I can answer your questions by walking through some plots. Take a look at this Wikipedia link on risk aversion. Go the section labeled 'Example' (or, just see the images I've pasted...

As a corollary, something rvince99 wrote earlier in this thread caught my attention. I haven't actually worked this out, but given logarithmic utility, it seems inconsistent to add to a losing position/widen a stop loss (this is when the loss of utility is greatest) or to add ('scale-in') to a...

Maybe start with Kelly's original paper and/or just think through what Kelly proposed: maximize expected logarithmic utility given known and independent events, probabilities of the events, payoffs, a loss limited to the bet,..., I'm probably leaving something out. Does this sound like your...

You might like what Aaron Brown has to say about 'risk ignition' in his book "Red-Blooded Risk'. He talks about maximizing position size when you think you have a trade with an unusual edge, constrained by a VaR limit. It's related to Vince's ideas.

Here's one of Vince's papers on the subject. https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2577782 I don't think Vince's argument is that Optimal f is superior, more that Kelly is inappropriate for many trades. For instance, short sales, instruments/trades with non-linear payoffs...

Update: I spoke with the owner of QuantTerminal (QT) yesterday after testing a demo version of the product. I liked what Daniel had to say and I plan on signing up for a license. He's a small business owner like many of us. I found him to be forthright and generous with his time. The...

First time I've seen anyone mention Chiang's book. I liked that one, had it in grad school.

ARPM is first-rate. I attended their week-long on-site program a few years ago. Risk probably employs more quants than the other areas. This is a good exposure to the requisite applied mathematics. Tho, IIRC, it's mostly geared toward people in industry, so you'll have a bit of a learning...