they aren't true residualized returns if they don't come from the idiosyncratique region of France

Pro tip: run a regression and then swap the coefficients with the p-values. I've seen this produce R^2 values significantly greater than 100%

In HFT most smoothing is with an EWMA because the devs get annoyed when you ask them to keep a long cache of prices and keep the system performant.

Volatility isn’t so bad when it only goes up

Here’s a quick thread on the essentials of a market making system that the pros use. …

Prediction markets are crypto (at least) a few years ago in terms of finding edges. Pay attention.

You can tell pretty fast when you meet someone if they love the game or they hear it’s a place to make a bit of money and have a nice salary. There’s a lot of people who want to break into this industry because it pays well. Speak to “aspiring investment bankers” - half of them don’t even know what you do on the job, but they know it pays. They’re trying to achieve, not truly win the top prize, not dominate. There’s those who clearly want to win and those who want a nice salary. The proof is in the pudding and you’ll see it in their track. You’ll meet people who didn’t ask for permission and just went and did things, and whilst to some that looks like high agency it’s not just about agency it’s about their unwillingness to sit by and take a standard path. It’s the aggression of their agency which reveals their love for the game and the urgency to win. On the same note those who are generalists, true generalists with little consistent theme who hop from latest trend to latest trend, have already shown their cards. Their sheer inability to stick with a field has already revealed a that they don’t truly love it. If you actually were engulfed by the work and the love of the game you would find it hard to move to the greener grass in the next industry and even then you’d get pulled right back. That’s not to say that people don’t do very well under other conditions, but it’s those who truly love the game that do a large majority of the winning. Wanting to make money will only carry you so far. Often, if you give the person who swears they’re motivated to be the best quant in the game a couple hundred grand a year and suddenly they care about job security. You see how much people want to truly win, and take home first prize, once people start winning because a lot of people really just want to achieve and for them achieving is a high paid job. Enough to be happy and having that amount come in often. Those who truly want to win and not much else in between separate themselves in their actions and their agency and a lot of those who claim to be part of this group are not.

Best advice I got from Steve Cohen: volatility is like a trampoline

Almost at 500 questions / answers on the blog Q&A. Lots of great reading there available to all, would recommend. Should be on sub-stack

When you read someone’s notebook and theres comments on every line