Files
ai/gateway/knowledge/trading/strategies/etfs/mean-reversion.md
Tim Olson 47471b7700 Expand model tag support: add GLM-5.1, simplify Anthropic IDs, scan tags anywhere in message
- Flink update_bars debouncing
- update_bars subscription idempotency bugfix
- Price decimal correction bugfix of previous commit
- Add GLM-5.1 model tag alongside renamed GLM-5
- Use short Anthropic model IDs (sonnet/haiku/opus) instead of full version strings
- Allow @tags anywhere in message content, not just at start
- Return hasOtherContent flag instead of trimmed rest string
- Only trigger greeting stream when tag has no other content
- Update workspace knowledge base references to platform/workspace and platform/shapes
- Hierarchical knowledge base catalog
- 151 Trading Strategies knowledge base articles
- Shapes knowledge base article
- MutateShapes tool instead of workspace patch
2026-04-28 15:05:15 -04:00

3.0 KiB

description, tags
description tags
Constructs a dollar-neutral ETF portfolio by selling ETFs with high Internal Bar Strength (IBS, close near daily high) and buying ETFs with low IBS (close near daily low), exploiting short-term mean-reversion.
etfs
mean-reversion
ibs
internal-bar-strength

Mean-Reversion (ETFs)

Section: 4.4 | Asset Class: ETFs | Type: Mean-Reversion

Overview

This strategy applies mean-reversion to ETFs using the Internal Bar Strength (IBS) indicator, derived from the previous day's close, high, and low prices. ETFs with a close near their daily high (high IBS) are considered "rich" and likely to revert downward; ETFs with a close near their daily low (low IBS) are "cheap" and likely to revert upward. A dollar-neutral portfolio sells high-IBS ETFs and buys low-IBS ETFs.

Construction / Signal

Internal Bar Strength (IBS):

IBS = (P_C - P_L) / (P_H - P_L)                           (370)

Where:

  • P_C = previous day's closing price
  • P_H = previous day's high price
  • P_L = previous day's low price

IBS ranges from 0 to 1:

  • IBS close to 1: price closed near the daily high → ETF is "rich"
  • IBS close to 0: price closed near the daily low → ETF is "cheap"

An equivalent symmetric measure: Y = IBS - 1/2 = (P_C - P_*) / (P_H - P_L) where P_* = (P_H + P_L) / 2; Y ranges from -1/2 to +1/2.

Portfolio construction:

  • Sort ETFs cross-sectionally by IBS.
  • Sell ETFs in the top decile (high IBS, "rich").
  • Buy ETFs in the bottom decile (low IBS, "cheap").
  • Dollar-neutral construction.

Entry / Exit Rules

  • Entry: Each day after the close, compute IBS for all ETFs, rank, and enter positions for the next day's open or close.
  • Exit: Typically hold for 1 day (short-term mean-reversion); close at next day's close.
  • Rebalance: Daily.

Key Parameters

  • IBS computation: Daily, using previous day's high, low, and close
  • Holding period: Short-term (typically 1 day)
  • Portfolio construction: Dollar-neutral long/short decile
  • Weights: Uniform for all long and all short ETFs, or volatility-weighted

Variations

  • Volatility-weighted positions: Weight positions by historical ETF volatility rather than equal-weighting
  • Stock mean-reversion methods: Mean-reversion strategies from Section 3 (cluster, weighted regression) can also be adapted to ETFs
  • IBS threshold: Instead of top/bottom decile, use a fixed IBS threshold (e.g., IBS > 0.8 = short, IBS < 0.2 = long)

Notes

  • IBS is a simple, daily-bar indicator requiring only OHLC (open-high-low-close) data.
  • Mean-reversion in ETFs can be stronger than in individual stocks because ETFs represent diversified baskets where idiosyncratic volatility is reduced, and market-maker arbitrage constrains large deviations from NAV.
  • Holding period is very short (1 day); transaction costs can be significant for daily rebalancing.
  • The strategy can be combined with other signals (e.g., sector momentum) for confirmation.
  • All stock-based mean-reversion strategies (clusters, weighted regression) can be adapted for ETF universes.