2.82 Regularization from First Principles
OLS is the best unbiased fit only under assumptions markets shatter, so on correlated alphas it hands you wild coefficients. Add an L2 or L1 penalty: biased, lower variance, steadier out of sample.
Author
ali askar
8.8 Itô Calculus, Intuitively
Stochastic calculus is one twist on the integral you know. Riemann weights rectangles by equal width, Riemann-Stieltjes by a function g, Itô by Brownian motion. Random weights make the integral a process Y(t).
7.3 Rust Data Server, Python Brain
Profiling a slow algo points at JSON decode, not your alpha. Put the simple hot path (decode, filter, forward) in Rust and keep the complex stateful brain (OMS, reporting, unwinding) in Python. Split at the JSON, forward binary, and match your backtest exactly.
7.2 Dynamic Time Warping for Time-Series Alignment
Fixed-lag cross-correlation assumes the delay between two markets never moves. It does. Dynamic time warping aligns two series with a stretchable path, scoring shape similarity and recovering a lead-lag that varies through time.
7.1 Synthetic Prices: EMA of Random Numbers as a Market Model
Real prices are random numbers with memory, so an EMA of noise plus a cumulative sum makes a passable synthetic chart. Generate thousands to put error bars on a backtest, but never to prove edge.
6.45 Don't Trust Your Strategy-Weighting Scheme
Scoring strategies four ways then averaging the weights isn't rigorous: the metrics collapse to two correlated families and every input is a noisy in-sample estimate. Beat equal-weight out of sample.
5.42 Why Market-Making Simulations Don't Work
You can't simulate a limit fill that never happened. Maker/taker on historical trades works only on thin venues; pure making is just prod tuning. And Avellaneda-Stoikov is risk times root holding time.
5.41 Quotes From the Future: Clock Drift and the Latency Floor
A quote stamped newer than light can travel isn't fast, your clocks disagree. Fix chrony first, then use an empirical latency floor (half the fastest ping) to catch and correct cross-venue drift.
3.36 The NATGAS 20-25h Cycle: Real or Folklore?
A NATGAS daytrader legend claims a guaranteed 20-25h sell-wave cycle, 44 trades a month. Run it through the gauntlet: cause test, multiple-comparisons, permutation null, and cost arithmetic. It is folklore until proven otherwise.
3.35 Trend and Reversion Are the Same, and OLS Understates Both
Trend and reversion are one process split by the sign of beta. Regress a series on its own lag and OLS understates beta's magnitude in both cases, because the lagged regressor shares error terms with the target. You end up believing in less trend and less reversion than the market carries.
2.81 Volume and Volatility Are the Same Feature
Replace the volatility term with volume in most alphas and the backtest barely moves, because they ride the same information clock. Feeding a model both is double-counting one factor. Keep one scale, and add their ratio, Amihud illiquidity, as the residual that actually carries new information.
2.80 Ridge Above 1h, XGBoost Below 5min
Pick the model by interaction strength: ridge when near-additive, XGBoost when interactions dominate. XGBoost also handles NaNs natively, avoiding imputation lookahead; else use an IC-weighted ensemble.