Rolling Sharpe Ratio

Quick Reference

Property	Value
Dimension	quality
Category	statistics
Version	v1.0
Output Column	sharpe_ratio

Rolling Sharpe ratio: rolling_mean(return) / rolling_std(return) - risk-adjusted return over window

Formula

rolling_mean(ret, window) / rolling_std(ret, window)

CDM Inputs

Column	CDM Table	Description
ret	cdm_*	CDM source table

Parameters

Parameter	Type	Default	Description
window	integer	300000	Window for mean and std estimation

Output

Column: sharpe_ratio

Rolling annualized Sharpe ratio

Market Intuition & Trading Rationale

Rolling Sharpe ratio measures risk-adjusted return: rolling_mean(ret) / rolling_std(ret). It answers: how much return am I getting per unit of risk over the recent window? A Sharpe of 1.0 means the strategy earns one unit of return per unit of volatility — decent. A Sharpe of 2.0+ is excellent. A negative Sharpe means the strategy is losing money on a risk-adjusted basis.

The rolling window makes this adaptive — it captures the strategy's current performance, not its lifetime average. A strategy with a long-term Sharpe of 1.5 may have experienced periods with rolling Sharpe of -2.0 (drawdowns) and +4.0 (strong runs). Monitoring the rolling Sharpe detects when a strategy's performance regime has changed — a rolling Sharpe crossing below zero is an early warning that the strategy's edge may have disappeared.

Usage Cases

• Strategy monitoring: Track rolling_sharpe_ratio as a live quality metric. When it drops below a threshold (e.g., 0.5) for an extended period, the strategy may be broken — pause live trading and investigate.
• Dynamic capital allocation: Allocate more capital to strategies with higher rolling Sharpe. This is a simple, robust risk-parity approach that automatically reduces exposure to underperforming strategies.
• quality dimension: Used in intraday_momentum as the quality assessment — the momentum signal's Sharpe over the recent window tells you whether momentum is currently a profitable factor in this instrument.

YAML Definition

name: rolling_sharpe_ratio
description: 'Rolling Sharpe ratio: rolling_mean(return) / rolling_std(return) - risk-adjusted
  return over window'
category: statistics
dimension: quality
version: v0.9.0 (Beta)
required_inputs:
  - ret
output_column: sharpe_ratio
output_description: Rolling annualized Sharpe ratio
tags:
  - quality
  - sharpe
  - risk
parameters:
  window:
    type: integer
    description: Window for mean and std estimation
    required: false
    default: 300000
formula: rolling_mean(ret, window) / rolling_std(ret, window)