Machine Learning Classification and Portfolio Allocation: with Implications from Machine Uncertainty
Fuente:
arXiv
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Preprint |
| Published: |
2021
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866916857305366528 |
|---|---|
| author | Bai, Yang Pukthuanthong, Kuntara |
| author_facet | Bai, Yang Pukthuanthong, Kuntara |
| contents | We use multi-class machine learning classifiers to identify the stocks that outperform or underperform other stocks. The resulting long-short portfolios achieve annual Sharpe ratios of 1.67 (value-weighted) and 3.35 (equal-weighted), with annual alphas ranging from 29\% to 48\%. These results persist after controlling for machine learning regressions and remain robust among large-cap stocks. Machine uncertainty, as measured by predicted probabilities, impairs the prediction performance. Stocks with higher machine uncertainty experience lower returns, particularly when human proxies of information uncertainty align with machine uncertainty. Consistent with the literature, such an effect is driven by the past underperformers. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2108_02283 |
| institution | arXiv |
| publishDate | 2021 |
| record_format | arxiv |
| spellingShingle | Machine Learning Classification and Portfolio Allocation: with Implications from Machine Uncertainty Bai, Yang Pukthuanthong, Kuntara General Finance Machine Learning General Economics Economics Computational Finance Portfolio Management We use multi-class machine learning classifiers to identify the stocks that outperform or underperform other stocks. The resulting long-short portfolios achieve annual Sharpe ratios of 1.67 (value-weighted) and 3.35 (equal-weighted), with annual alphas ranging from 29\% to 48\%. These results persist after controlling for machine learning regressions and remain robust among large-cap stocks. Machine uncertainty, as measured by predicted probabilities, impairs the prediction performance. Stocks with higher machine uncertainty experience lower returns, particularly when human proxies of information uncertainty align with machine uncertainty. Consistent with the literature, such an effect is driven by the past underperformers. |
| title | Machine Learning Classification and Portfolio Allocation: with Implications from Machine Uncertainty |
| topic | General Finance Machine Learning General Economics Economics Computational Finance Portfolio Management |
| url | https://arxiv.org/abs/2108.02283 |