Quant Beckman

🟥For this one and more papers with code: quantbeckman.com

The paper admits that the baseline uses the last observed return rather than a true terminal delisting payoff, and its own sensitivity check shows the bias rises from 4.94pp to 5.2pp, 5.8pp, or 6.5pp under harsher delisting assumptions.

@chickenfender ahahhaahah the epic win for the investors

@quantbeckman Improve your models with a this one simple trick: test on the training data

They say the model is trained from July 1999 to July 2016, but then they backtest it on the whole dataset from July 1999 to July 2020⁉️

EMH is the antithesis of inefficiency. Are there still people who consider it?

MEMM picks the pricing measure Q* closest to the historical measure P in relative-entropy geometry. This makes the measure change minimal while enforcing the martingale constraint for discounted prices.

@HackworthAI 😏

@quantbeckman all possible parameterizations and/or optimal controls?

First: Reduced model --> Then: Full model

🟥For this one and more papers with code: quantbeckman.com

It looks like ListFold-exp still needs more investigation, but it’s pretty interesting

What is the point of building something that has no validity or meaning?

These conditionals are points on a probability manifold, so edge persistence becomes a geometric closeness test across time. When the conditional law for the edge state moves toward the baseline or toward other states, the edge decays.

🟥For this one and more papers with code: quantbeckman.com

@benjmttt A NNet aproximating noise

@quantbeckman The ones who survived treated the LLM as a signal preprocessor not an alpha generator. Different architecture, different expectations, different results. What was the specific breakdown you saw most?

Academy VS reality. This is how the LLM #trading season has ended.

@TheSignlProcess 💯

The compression framing is the useful part. I think most signals fail not because they lack predictive power in isolation but because they discard information that only becomes relevant when the market structure shifts. A sufficient statistic for a stationary market isn’t sufficient anymore when the regime changes.

📘[QUANT LECTURE] Sufficient statistics and minimal signals📘

@theone34824249 There is always something better

@quantbeckman Is there a better model than equations with just reasoning ???

Mmmm since the study uses an unlevered long-only framework over 1991–2020, part of the strong MPO RP result may reflect a bond-friendly era.