Cross-Sectional and Time-Series Momentum Compared
Summary
The document distinguishes cross-sectional momentum (CSM), which ranks assets against peers, from time-series momentum (TSM), which measures each asset’s own absolute performance. The discussion presents CSM as useful for isolating relative, asset-specific strength or weakness, while TSM can reflect a broad market rise or fall shared by many assets. It frames the choice as a question about whether relative outperformance or an asset’s own trend better signals continuation.
The answer favors CSM for a stock momentum factor and attributes that view to a portfolio management text, but provides no empirical tests or comparative performance evidence. It also raises possible drivers such as investor redemptions, post-earnings drift, and herding without resolving their role. The distinction is conceptual: neither signal is shown to be universally superior, and the discussion does not specify portfolio construction, lookback periods, or risk controls.
Key ideas
- Cross-sectional momentum ranks an asset’s performance relative to other assets.
- Time-series momentum evaluates an asset’s absolute performance over time.
- Time-series signals can reflect market-wide movement when many assets rise or fall together.
- Cross-sectional momentum is presented as a way to focus on relative, potentially idiosyncratic performance.
- The document raises behavioral and event-related explanations but does not test them.
Tags
Full text
# Cross sectional momentum vs time series momentum # Cross sectional momentum vs time series momentum What are the advantages/disadvantages of creating quantitative strategies using cross sectional momentum vs time series momentum? From my perspective, time series momentum is a better indicator of continued increase/decrease as it indicates absolute performance of the stock. Although, at the same time, perhaps cross sectional momentum is better because if a stock is outperforming/underperforming similar stocks in the market then I would think it more strongly be pushed by the market to continue in its same direction because of its uniqueness? In other words, if it performs better RELATIVE to other stocks then the market will take notice of that more strongly. Also, is the profitability of both forms of momentum subject a result of things like redemption requests, post-earnings announcement drift, herding effect, etc? I just learned about this so sorry if my question doesn't make sense, thank you! ## Answer by KaiSqDist (score 0, accepted) https://quant.stackexchange.com/a/79716 In "Advanced Portfolio Management", Paleologo is an advocate of CSM, I agree that the momentum of stocks relative to others is a much better indicator for the momentum factor. Because if you just take TSM, it could be possible that all stocks are on the rise/fall at that point in time, which just identifies the momentum of the market and not the idiosyncratic (CSM) momentum. UPDATE I would like to quote this line from the book (section 5.2.3, page 62), which I think resolves this dilemma well - "Momentum has a different meaning from trend-following, because it is based on relative performance, whereas the latter is based on absolute performance".
Shown in full with attribution under the source's licence. Licence: CC BY-SA 4.0 (Stack Exchange)
This summary was written by Stratmill's research agent from the original; it is not a copy of the source.