# Expert Aggregation for Financial Forecasting

arXiv: https://arxiv.org/abs/2111.15365 · HTML: https://mlsanity.com/paper/2111.15365/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- Builds on 6 of its 64 references that are in the map.
- First posted on arXiv 25 Nov 2021.

## Builds on (most cited first, 6 of 6)

- [A Second-order Bound with Excess Losses](https://mlsanity.com/paper/1402.2044/) (2014)
- [Relative Loss Bounds for On-line Density Estimation with the Exponential Family of Distributions](https://mlsanity.com/paper/1301.6677/) (2013)
- [Forecasting: theory and practice](https://mlsanity.com/paper/2012.03854/) (2020)
- [Optimal learning with Bernstein Online Aggregation](https://mlsanity.com/paper/1404.1356/) (2014)
- [Learning Multiple Stock Trading Patterns with Temporal Routing Adaptor and Optimal Transport](https://mlsanity.com/paper/2106.12950/) (2021)
- [Efficient tracking of a growing number of experts](https://mlsanity.com/paper/1708.09811/) (2017)

## Built on by (most cited first, 3 of 3)

- [Adaptive Conformal Predictions for Time Series](https://mlsanity.com/paper/2202.07282/) (2022)
- [Forecasting time series with constraints](https://mlsanity.com/paper/2502.10485/) (2025)
- [MoFE: A Novel Mixture-of-Experts Framework with Fourier Neural Operators for Cryptocurrency Forecasting](https://mlsanity.com/paper/2608.17342/) (2026)

## Abstract (arXiv)

Machine learning algorithms dedicated to financial time series forecasting have gained a lot of interest. But choosing between several algorithms can be challenging, as their estimation accuracy may be unstable over time. Online aggregation of experts combine the forecasts of a finite set of models in a single approach without making any assumption about the models. In this paper, a Bernstein Online Aggregation (BOA) procedure is applied to the construction of long-short strategies built from individual stock return forecasts coming from different machine learning models. The online mixture of experts leads to attractive portfolio performances even in environments characterised by non-stationarity. The aggregation outperforms individual algorithms, offering a higher portfolio Sharpe Ratio, lower shortfall, with a similar turnover. Extensions to expert and aggregation specialisations are also proposed to improve the overall mixture on a family of portfolio evaluation metrics.