Simple Jump Indicator

Quick Reference

Property	Value
Dimension	signal
Category	volatility
Version	v1.0
Output Column	jump_indicator

Simple jump indicator: abs(return) > threshold * rolling_std(return) 鈥?detects extreme price moves

Formula

abs(ret) > (rolling_std(ret, window) * threshold)

CDM Inputs

Column	CDM Table	Description
ret	cdm_trade_enriched	Trade data enriched with bar context 鈥?price, volume, side, trade type

Parameters

Parameter	Type	Default	Description
threshold	float [1.0, 10.0]	3.0	Number of standard deviations for jump threshold
window	integer	60000	Window for rolling standard deviation in milliseconds

Output

Column: jump_indicator

Binary jump detection flag (0 or 1)

Market Intuition & Trading Rationale

Simple jump indicator is a binary flag: |return| > k 脳 rolling_std(return). When a return exceeds k standard deviations of recent volatility, a jump is detected. Unlike the continuous jump_volatility_detector, this outputs a simple 0/1 flag per observation 鈥?clean, easy to use as a filter. The default k=3 catches only extreme moves (~0.3% of observations under normality).

Usage Cases

• Event filtering: Exclude jump observations from volatility estimation to get cleaner, diffusive-only vol estimates.
• Trade suspension: Pause trading when jump_indicator fires 鈥?the market is experiencing an extreme event, and normal signal logic may not apply.
• context: Used in jump_microstructure_noise pack alongside microstructure_noise_ratio.

YAML Definition

name: simple_jump_indicator
description: 'Simple jump indicator: abs(return) > threshold * rolling_std(return)
  鈥?detects extreme price moves'
category: volatility
dimension: signal
version: v0.9.0 (Beta)
required_inputs:
  - ret
output_column: jump_indicator
output_description: Binary jump detection flag (0 or 1)
tags:
  - jump
  - volatility
  - signal
parameters:
  threshold:
    type: float
    description: Number of standard deviations for jump threshold
    required: false
    default: 3.0
    constraints:
      min: 1.0
      max: 10.0
  window:
    type: integer
    description: Window for rolling standard deviation in milliseconds
    required: false
    default: 60000
formula: abs(ret) > (rolling_std(ret, window) * threshold)