← All research
Research
Experiments
My notebook. Every note is one test of a handful of raw market measurements against one target on one dataset, written up the same way every time: the question, the data, the result, pass or fail. A measurement has to predict something on its own before a network is allowed to learn from it. This is where new ideas either earn a place in a system or get dropped.
Reading order
Experiment01
Testing four raw order-book measurements: three passed
Three passed. One dropped. The data picked the seeds.
Experiment02
Four ways to measure trade aggression, and three were the same thing
Queue-normalised size, sweep depth, consumed fraction, through-ticks. Only one is its own axis.
Sample draft03
When two signals tell the same story
A proposed redundancy check before adding another feature to a research model.
Sample draft04
What happens after a burst of trades?
A study outline for separating a sudden increase in trading activity from a directional prediction.
The course, and other research