A signal using an earnings release at 08:00 cannot produce an order at 07:59, and a signal using the official closing price cannot assume an execution at that same price after the auction has finished. Timing is a condition of an executable trading policy. This lesson implements the clock discipline underlying a point-in-time dataset, then connects usable inputs to decisions, orders, fills and labels.
Learning objectives. By the end of this note you should be able to:
- distinguish economic reference time, event time, publication time and receipt time;
- derive the earliest feasible decision from several delayed inputs;
- detect same-bar leakage and revised-record timing errors;
- specify timezone, clock-ordering and missing-timestamp conventions;
- design a small timing audit that fails on impossible decisions.
In this lesson
- A timestamp names an event
- Compose delays rather than assume them away
- Daily bars hide the same problem
- Make the audit executable
- Calendar and receipt conventions
- Timing continues through the research process
Before you start
The full lesson, with worked examples and exercise solutions, is for members.
Read the full lessonNew to SelfMadeQuant.com? See membership options or browse the Quantitative Research, Data and Backtesting syllabus.