research.morphiqlabs.com
Working papers and engineering write-ups from MorphIQ Labs on the Ferro suite, quantitative research, and Anvil. Market microstructure, volatility modeling, regime detection, and the runtime underneath.
Market-regime research is unusually easy to make convincing after the fact. Choose a familiar period, draw colored bands behind a price chart, and tell a story about what the model saw. A chart can look plausible even…
A regime label is a compressed description of recent market behavior. It is not a forecast, and it is not a trade.
SWT, synchrosqueezing, EMD, VMD, and EWT are often placed in the same comparison table because they all decompose a signal. That makes them look like interchangeable candidates for one job. They are not.
A research result and a production system are two different artifacts, and the distance between them is where correctness quietly fails. A pricing model is validated in a notebook and re-implemented in a fast language…
The most expensive production bugs in trading systems are often not the ones that crash. They are the ones that cannot be reproduced. A signal changed, an order decision differed, a timeout fired in a different order,…
Most options tools start with a table. Rows of contracts, columns of Greeks, filters, probabilities, and implied volatility fields. The table is useful, but it is not a decision process. A trader still has to assemble…
Price-time priority sounds simple until the implementation has to survive real state transitions. Orders arrive, cancel, partially fill, cross multiple price levels, and interact with timestamps, client identifiers, a…
Market-data infrastructure becomes fragile when every downstream component learns a different venue dialect. One feed calls it a trade condition, another calls it a flag. One stream uses exchange timestamps, another c…
A Rust options pricing engine usually gets judged by surface area first: how many models, how many Greeks, how fast the implied-volatility solver runs. That is necessary, but it is not enough. The harder question is w…
Time-series ML fails quietly when the feature layer ignores scale. A price series, volatility series, or sensor stream may look like one signal, but its behavior is usually a mixture of fast noise, medium-horizon stru…
Most trading systems are tested the way most software is tested: feed an input, assert an output, hope the parts you did not assert on behaved. That works until the part you did not assert on is wall-clock time, or it…
An implied-volatility surface is a deceptively small object: a function from log-moneyness and maturity to a single number. The trouble is that not every such function corresponds to a market. A surface that prices a …
"The market changed" is the kind of statement that is obviously true and almost useless. Changed how fast? Changed in the minute-scale chop, or in the multi-day trend? A single rolling volatility number cannot tell yo…