Ma Crossover

Quick Reference

Property	Value
Dimension	technical
Category	technical
Version	v0.9.0 (Beta)
Output Column	ma_crossover

MA crossover signal: fast_ma > slow_ma - binary trend signal when faster moving average crosses above/below slower MA

Formula

diff((rolling_mean(price, fast_window) > rolling_mean(price, slow_window)), 1)

CDM Inputs

Column	CDM Table	Description
price	cdm_*	CDM source table

Parameters

Parameter	Type	Default	Description
fast_window	integer [2, 200]	10	Fast moving average window
slow_window	integer [5, 1000]	50	Slow moving average window

Output

Column: ma_crossover

Crossover signal: 1=fast above slow (bullish), 0=fast below slow (bearish)

Market Intuition & Trading Rationale

MA Crossover produces a binary trend signal: diff(fast_ma > slow_ma, 1). When the fast moving average crosses above the slow MA, the signal pulses +1 (bullish crossover / "golden cross" when using 50/200). When it crosses below, -1 (bearish crossover / "death cross"). Most of the time the signal is 0 — crossovers are rare events that mark potential trend changes.

The diff-of-comparison construction is elegant: fast_ma > slow_ma produces a boolean (1 when fast is above, 0 when below), and diff(..., 1) converts level changes into pulses. A transition from 0 to 1 produces +1 (just crossed above). 1 to 0 produces -1 (just crossed below). No change produces 0. This gives you a clean event signal — fire on crossover, stay flat the rest of the time.

The default windows (10/50) capture short-term vs medium-term trend alignment. Smaller windows (5/20) produce more frequent crossovers suitable for intraday trading. Larger windows (50/200) produce the classic "golden cross/death cross" signals used for long-term trend following.

Usage Cases

• Trend following entries: Go long on +1 crossover, go short on -1 crossover. The crossover confirms that the short-term trend has aligned with (or broken away from) the medium-term trend.
• Trend filter for other strategies: Only take long signals from other features when ma_crossover > 0 (fast above slow = uptrend). Only take short signals when ma_crossover < 0. This simple filter eliminates counter-trend trades.
• Regime classification: ma_crossover = 1 means trending up; = -1 means trending down; and the duration since the last crossover tells you how mature the trend is. Recently crossed → early trend (higher potential). Crossed long ago → mature trend (closer to exhaustion).

YAML Definition

name: ma_crossover
description: 'MA crossover signal: fast_ma > slow_ma - binary trend signal when faster
  moving average crosses above/below slower MA'
category: technical
version: v0.9.0 (Beta)
dimension: technical
status: Pre-release
required_inputs:
  - price
output_column: ma_crossover
output_description: 'Crossover signal: 1=fast above slow (bullish), 0=fast below slow
  (bearish)'
parameters:
  fast_window:
    type: integer
    description: Fast moving average window
    required: false
    default: 10
    constraints:
      min: 2
      max: 200
  slow_window:
    type: integer
    description: Slow moving average window
    required: false
    default: 50
    constraints:
      min: 5
      max: 1000
formula: diff((rolling_mean(price, fast_window) > rolling_mean(price, slow_window)),
  1)