When Harry met Kelly: an overlooked result in the classical theory of optimal capital growth

Richard Watt

Annals of Finance2026https://doi.org/10.1007/s10436-026-00477-0article
AJG 2ABDC B
Weight
0.50

Abstract

A classic problem in finance is the question of how an investor should optimally allocate wealth over one risky and one risk-free asset. This article reconsiders this theory under the assumptions that (1) the investor’s objective is to maximize the intertemporal growth rate of their wealth, (2) that the investor can make portfolio adjustments only in discrete time, and (3) that the risky asset is defined in each period according to only its mean return and the standard deviation thereof. This problem has been resolved only for continuous time. Here, the discrete time strategy for optimal capital growth is resolved for two different methodologies, and it is shown that the continuous time model is recoverable as a limit version of each of the two discrete time models. Finally, both of the methodologies considered lead to the same first-order Taylor’s approximation formula, which always gives a more accurate approximation to the true optimal discrete time strategy than does the continuous time strategy.

Open via your library →

Cite this paper

https://doi.org/https://doi.org/10.1007/s10436-026-00477-0

Or copy a formatted citation

@article{richard2026,
  title        = {{When Harry met Kelly: an overlooked result in the classical theory of optimal capital growth}},
  author       = {Richard Watt},
  journal      = {Annals of Finance},
  year         = {2026},
  doi          = {https://doi.org/https://doi.org/10.1007/s10436-026-00477-0},
}

Paste directly into BibTeX, Zotero, or your reference manager.

Flag this paper

When Harry met Kelly: an overlooked result in the classical theory of optimal capital growth

Flags are reviewed by the Arbiter methodology team within 5 business days.


Evidence weight

0.50

Balanced mode · F 0.40 / M 0.15 / V 0.05 / R 0.40

F · citation impact0.50 × 0.4 = 0.20
M · momentum0.50 × 0.15 = 0.07
V · venue signal0.50 × 0.05 = 0.03
R · text relevance †0.50 × 0.4 = 0.20

† Text relevance is estimated at 0.50 on the detail page — for your query’s actual relevance score, open this paper from a search result.