Files
ai/gateway/knowledge/trading/strategies/etfs/sector-momentum-rotation.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

3.8 KiB
Raw Blame History

description, tags
description tags
Overweights ETFs from outperforming sectors and underweights those from underperforming sectors based on T-month cumulative return momentum, with optional MA filter and dual-momentum variants.
etfs
momentum
sector-rotation

Sector Momentum Rotation

Section: 4.1 / 4.1.1 / 4.1.2 | Asset Class: ETFs | Type: Momentum / Sector Rotation

Overview

Empirical evidence shows that the momentum effect exists not only for individual stocks but also for sectors and industries. The sector momentum rotation strategy overweights ETFs from outperforming sectors and underweights those from underperforming sectors, using ETFs concentrated in specific sectors/industries to implement sector/industry rotation without buying or selling large numbers of underlying stocks.

Construction / Signal

Similarly to stock price-momentum (Section 3.1), use each sector ETF's cumulative return as the momentum measure. Let P_i(t) be the price of ETF labeled by i:

R_i^cum(t) = P_i(t) / P_i(t + T) - 1                     (361)

Here t + T is T months in the past w.r.t. t. After time t, buy ETFs in the top decile by R_i^cum(t) and hold for a holding period (typically 13 months).

Dollar-neutral construction: Buy top-decile ETFs and short bottom-decile ETFs (ETFs can be shorted).

Long-only construction: Buy only top-decile ETFs, equal-weight or volatility-weight.

Entry / Exit Rules

  • Entry: At rebalance, rank all sector ETFs by cumulative return R_i^cum; buy top-decile, optionally short bottom-decile.
  • Exit: Hold for 13 months; rebalance at the next scheduled interval.
  • Formation period T: Typically 612 months.

Key Parameters

  • Formation period T: 612 months
  • Holding period: 13 months
  • Portfolio construction: Long-only (top decile) or dollar-neutral (top long, bottom short)
  • Weights: Uniform or volatility-adjusted

Variations

4.1.1 — Sector Momentum Rotation with MA Filter

A refinement that requires an ETF to pass a moving average filter before entering a position, preventing buys in sectors with downward price trends even if they rank high by relative momentum.

Rule = { Buy top-decile ETFs only if P > MA(T')
        { Short bottom-decile ETFs only if P < MA(T')       (362)
  • P = ETF's current price at transaction time
  • MA(T') = moving average of ETF's daily prices over T' days (T' can differ from formation period T; typically T' = 100200 days)

This ensures the absolute price level (trend) also supports the trade direction.

4.1.2 — Dual-Momentum Sector Rotation

In long-only strategies, mitigates the risk of buying sector ETFs when the broad market is trending down. Augments relative (cross-sectional) momentum with absolute (time-series) momentum of a broad market index ETF:

Rule = { Buy top-decile ETFs if broad market P > MA(T')
        { Buy an uncorrelated ETF (e.g., gold, Treasury) if broad market P <= MA(T')   (363)
  • P = broad market index ETF's price at transaction time
  • MA(T') = moving average of the broad market index ETF's price; typically T' = 100200 days

If the broad market is below its moving average (downtrend), capital is rotated into an ETF uncorrelated with the broad market (e.g., gold or Treasury ETF) instead of sector ETFs.

Reference: Antonacci (2014, 2017).

Notes

  • ETF-based sector rotation is simpler to implement than stock-level sector rotation: one ETF trade per sector instead of dozens of stock trades.
  • The MA filter (4.1.1) reduces the chance of buying momentum in a sector that is in absolute decline.
  • Dual-momentum (4.1.2) addresses the long-only strategy's vulnerability to broad market drawdowns.
  • Typical formation period: 612 months; typical holding period: 13 months.
  • Dollar-neutral construction removes broad market exposure but requires shorting ETFs (feasible in practice).