The Frontier Fund

The fund will be named the ‘Frontier Fund’, after the efficient frontier in the Markowitz model of portfolio optimisation.

The Markowitz model formalises the principle that investors should seek to maximise expected return for a given level of risk, or equivalently minimise risk for a given level of return. Portfolios that lie on the efficient frontier represent the mostefficient trade-offs between risk and return available within a given investment universe.

While the model has limitations—it is a single-period framework that considers only mean and variance and reduces riskto the market factor—it remains a valuable educational tool. Most importantly, it highlights that diversification is not simply about reducing risk, but about being appropriately compensated for the risks undertaken.

The Fund applies this framework in practice through the use of index ETFs, which help move the portfolio closer to the efficient frontier and create opportunities for experimentation with factor exposures (e.g. Size, Value, Momentum).

Financial engineering-type ETFs (e.g. leveraged funds, covered call funds) may make questionable investments due to high fees and issues like volatility decay (leveraged) and challenges in assessing skewness risk (covered calls). Thematic ETFs still contain a lot of uncompensated risk and are not as useful for moving toward the efficient frontier.

The fund uses real money primarily to explore the role of market microstructure in investing (particularly the presence of algorithmic traders, as well as issues like price impact, iceberg orders, execution, brokerage, etcetera), as these issues are typically not handled well in simulations.