Moneda Bench, made by Moneda

Moneda PortfolioBench

Decision quality on hypothetical portfolios: rebalancing, risk measures, position sizing and constraints; calibrated forecasts on dated events; and running a paper fund to a written mandate.

First results

Coming

About this benchmark

Size
150 portfolio tasks, about 100 forecasts a quarter, and mandate funds
Public split
Yes; forecasts resolve after they are asked, so there is nothing to memorise
Grading
Code for portfolio tasks. Brier score and calibration for forecasts. For mandate funds: how closely the fund follows its mandate, then costs, then risk-adjusted results against a simple rules-based portfolio.
Licence
Items CC BY 4.0, harness MIT
Submission
Open harness at release
Products allowed
No
What it does not cover
Investment performance. Results describe a model's decisions in a test setting with hypothetical money. Moneda does not manage portfolios.

How it will run

Portfolio tasks each have one right answer, so market noise plays no part. Forecasts are scored once events resolve. Mandate funds run first on a replay with no look-ahead, then for at least six months on live data, and are judged first on following the mandate, because short-run returns cannot separate skill from luck. Models only in public.

Status: first results coming.

Results

First results coming.

Paper, data and harness will be linked here at release.

Last checked 1 Nov 2026. Something out of date or wrong? Send a correction.