Momentum and network learning

The paper “Network Momentum across Asset Classes” introduces a multi-asset quantitative trading strategy based on network momentum—a trading signal derived from momentum spillover across different asset classes. Momentum spillover occurs when past performance in one asset predicts future returns in linked assets. While traditionally studied in pairwise connections (e.g., stock-bond pairs or supply chains), this paper expands the concept to a systemic network across four distinct asset classes.

The study analyzes daily price data from 1990 to 2022 across 64 continuous futures contracts covering Commodities, Equities, Fixed Income (FI), and Foreign Exchange (FX).

Methodology

Graph Learning - Uses a linear and interpretable graph learning model to infer dynamic asset networks strictly from historical price data, overcoming the lack of explicit fundamental links between disparate asset classes.

Momentum Features - Leverages 8 individual momentum features per asset, including volatility-scaled returns over multiple horizons (1-day to 1-year) and normalized Moving Average Convergence Divergence (MACD) indicators.

Portfolio Construction - Combines learned graph adjacency matrices with a linear regression model to forecast 1-day volatility-scaled returns across all assets.

Out-of-sample backtesting from 2000 to 2022 yields an annualized return of 22% and a Sharpe ratio of 1.51 after volatility scaling. The strategy exhibits low downside risk and low correlation with traditional individual momentum strategies. Inter-class (cross-asset) connections significantly enhance return predictability and diversification.

Treasury convenience yield and inflation

 

The convenience yield associated with Treasury securities is dynamic, meaning the price of safety associated with this safe asset is constantly changing with the macro environment. This important paper, "Inflation and Treasury Convenience,” on the macro dynamics of the convenience yield finds that inflationary supply shocks raise the opportunity cost of holding money and money-like assets, increasing convenience yields. Exogenous liquidity demand shocks will also elevate convenience but depress consumption and inflation. Given the difference between supply and demand shocks, there will be a weaker convenience-inflation link in the post-2000 period, which saw more liquidity demand shocks.  

This shows that convenience yield will be associated with macro dynamics and not just the demand for safety. My view is that this work makes it more difficult to discuss when there is a change in safety for Treasury assets. Yes, we can say it will be linked with macro dynamics, but ultimately most are interested in the price of safety based on some form of risk.

No momentum factor after accounting for cross-sectional liquidity

What causes momentum, or what is associated with momentum? A new paper suggests that there is a strong link between liquidity and momentum, and that changes in liquidity precede momentum gains.

The paper, titled “Momentum Returns and the Role of Liquidity Improvements” by Jeppe Bro, demonstrates that the traditional stock market momentum anomaly is actually driven by cross-sectional liquidity dynamics rather than representing an independent risk premium. 

The novel idea is that past winners systematically see their trading liquidity improve before a portfolio is formed, while losers see liquidity deterioration. The momentum price drift is the market adjusting to these new liquidity states. The author calls this the Liquidity Improvement Factor, and when including this factor, there is no momentum alpha. The data shows mixed results pre-2000 data relative to more recent data. Any momentum effect is drift toward high-liquidity stocks and is a byproduct of liquidity dynamics. This is the most recent paper that attempts to explain the momentum factor. 

This is a very interesting thesis. Investors should track or follow liquidity changes to enhance any measure of momentum. I have some issues with the Amihud measure of liquidity, which looks at absolute return divided by vol and is manipulated to form liquidity differences. Still, I do not have a better alternative at this time. 

Periods of financial stress - the long history

 

A paper that has not received much attention focuses on measures of systemic risk in "Systemic Risk Measures: From the Panics of 1907 to the Banking Stress of 2023". Much of the data from this paper is available from V-lab. The work suggests that there have been more stress periods recently, but their duration has been shorter. There has not been a stress period since the banking crisis of 2023. More importantly, the work examines what happens cross-sectionally to firms when they enter periods of stress and identifies specific institutions that face risk. 

The stress measures focus on US financial firms' stock return comovements and can predict market outcomes like realized volatility and returns, balance sheet outcomes, and bank failures. The stress periods focus on the contribution and exposure versions of CoVaR, marginal expected shortfall MES, and SRISK. 

Stress periods are identified through a two-step process based on narrative analysis, with start and end periods associated with the GZ credit spreads for more recent periods. 

The key finding is that market-based indicators offer distinct information that complements traditional balance sheet metrics for risk assessment.