cool kids these days use AGI to find HFT features but I'm just a washed unc using moving averages to sell vol
back when unc was interning at RenTech's vol desk, Jim Simons told unc "if you sell vol, it's ok to blow up tomorrow. just don't blow up today"
no crying in the casino
what kind of Sharpe & PnL does one need to generate before they can stop themselves from picking up and eating chicken tandoori that fell on the floor? asking for a friend
they say AGI is gonna unlock trillions of GDP but wait until you see normies using frontier model GPT-6 pro ™️ to ask for directions and food recommendations
For those who still didn't understand what Jev from @typesafeai can do, here is the example.
Jev can read dozens of structured trading signals and then show intelligent decisions.
It's time to make your system actually alive.
What a time!
guy joined training for the first time & he brought his son to watch so in sparring I let him work bcs I thought it’d be cool for the kid to see his dad crushing it
but now the kid looks at me w disgust like I’m some bitch ass that got his ass whooped I might need to change gym
@lonestar799@0xfdf@0xFaust12@orrdavid there are only 20! possible orderings so it’s fairly trivial I just try all of them and select the one with the best out-of-sample R2
@0xfdf@0xFaust12@orrdavid hi, regression expert here. 20 variables is obviously overfitted. but if I *have* to do it, I’ll start with one feature, residualise both y and the second feature against it, regress the residuals on each other, then repeat to 20. this is how we practitioners avoid overfitting