Portfolio Neutrality and Alternative Position Sizing Methods
Summary
The document surveys alternatives to Kelly sizing, focusing on how statistical arbitrage portfolios balance positions. Sector neutrality groups assets by industry, while risk neutrality uses a model of shared factor exposures to help an optimizer construct a balanced portfolio. In both cases, position sizes reflect relationships among assets as well as their standalone return prospects. The goal is to manage portfolio exposures, rather than maximize each position's expected return independently.
The discussion notes practical constraints that can keep a portfolio from reaching its ideal neutral exposures, including position and volume limits, short-sale locate availability, and transaction costs. It also names several other money management approaches: models catalogued by Van Tharp, Ralph Vince's Leverage Space Portfolio Model, and Ryan Jones' Fixed Ratio strategy. These are references rather than detailed explanations or comparisons; the document provides no performance evidence or guidance for choosing among them. It therefore serves as an overview of alternatives, not a recommendation that any one method is superior to Kelly or fractional Kelly.
Key ideas
- Sector neutrality balances portfolio holdings by industry affiliation.
- Risk neutrality uses factor exposures to guide portfolio construction.
- Trading constraints can prevent ideal neutral positions from being implemented.
- The document names several position sizing frameworks but does not compare their performance.
Tags
Full text
# Alternate money management strategies to Kelly? # Alternate money management strategies to Kelly? Other than Kelly (or fractional derivatives), are there other money management strategies in wide use among quant funds? Certainly Kelly is mathematically optimal, but perhaps there are other approaches that take into account factors beyond optimized return (lower risk, less exposure to X, etc.)? ## Answer by chrisaycock (score 7, accepted) https://quant.stackexchange.com/a/1321 In statistical arbitrage, quant traders attempt to build a neutral portfolio by balancing various assets against each other. Each asset's size within the portfolio isn't determine necessarily by how much money it's expected to generate, but by how correlated it is against other assets. A simple approach is sector neutrality, in which sector/industry affiliation is the sole criteria for "correlation". A better approach is risk neutrality, in which a risk model that describes each asset's exposure to common factors is fed into a portfolio optimizer. In either case, the goal is to build a portfolio where the assets balance each other. Neutrality is just a goal, though. There are other issues that may prevent a trader from obtaining the ideal positions, such as position limits, volume limits, inability to get the locates on a short sell, transaction costs, etc. The trading engine must consider these issues as well. ## Answer by babelproofreader (score 11) https://quant.stackexchange.com/a/1322 Van Tharp, in his book Definitive Guide to Position Sizing, identifies 31 separate models for money management. In said book he specifically warns against using both the Kelly Criterion and Optimal f. In addition to the models identified by Van Tharp there is Ralph Vince's Leverage Space Portfolio Model. ## Answer by Juan M. Almodóvar (score 0) https://quant.stackexchange.com/a/1396 Also Ryan Jones' Fixed Ratio strategy, from his book The Trading Game.
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.