Setting total return targets (TR) is tempting. It sells because humans hate variance but love targets. A target return looks like control. Combining target returns with a confidence corridor, that is, a promise that annual excess returns will stay within a narrow band (e.g., +/- 3%) around the target (e.g., 4%) most of the time, looks like discipline. A calendar year looks like a horizon. None of these are economics; they are psychology and reporting conventions, conflating measurement with control. The goal of this short note is not to moralize about optimism. It is to do the unpleasant arithmetic that converts a verbal TR objective into an implied Sharpe ratio requirement. Once that conversion is made, most "TR" claims can be recognized for what they imply: an attempt to smuggle a hedge-fund-like promise into a long-only balanced wrapper.
The Markowitz portfolio framework is widely used to determine static asset weights, while Merton's dynamic approach allows allocations to adjust with changing market conditions but is mathematically challenging and less practical. We address this gap by applying machine learning to dynamic portfolio optimization in the spirit of Merton, incorporating economic regimes defined by the VIX volatility index. An artificial neural network is trained to learn optimal allocation policies across regime-switching environments and is compared with classical regime-agnostic and theoretical regime-switching Merton strategies. On synthetic data with realistic constraints prohibiting borrowing and short selling, the machine learning strategy outperforms traditional benchmarks. Two empirical backtests-using monthly data from 1990 to 2025 and annual data from 1928 to 2025-show that accounting for regimes enhances performance and robustness.
