Data · dataset · 2026
Long-Flat Swing Trading Benchmark Strategy for BOVA11 ETF: a proposal
Listed in FGV Dataverse
Machine learning methods are now standard in empirical finance, often reporting economically meaningful return predictability.
Description
However, typical evaluations overstate implementable performance: trading costs are frequently ignored, basic risk controls such as stop-losses are rarely included, and extensive specification searches raise data-snooping and backtest-overfitting concerns. Predictive accuracy therefore differs from achievable returns.
We use this dataset in our research which aims to provide an easy-to-implement, transparent and reproducible empirical benchmark. We incorporate transaction costs and basic risk controls into a transparent pipeline for developing, backtesting and statistically evaluating a deliberately simple, medium-frequency swing trading strategy for the Brazilian ETF BOVA11, which tracks the Ibovespa index, at a ten-trading-day horizon.
Read the rest (1 more)
The benchmark is passive Buy-and-Hold, and the objective is to at least match it while incurring lower risk. We present the pipeline not as a novel trading signal, but as a reproducible benchmark for fairly evaluating more complex ML trading systems, reducing the gap between strategy modeling and real trading operations.
Links
Where it is published
- Dataverse dataset page dataverse.fgv.br/dataset.xhtml?persistentId=hdl%3A10438.3%2FFK2%2FPTGSZA ↗
landing page · from dataverse fgv br
- Persistent identifier hdl.handle.net/10438.3/FK2/PTGSZA ↗
DOI / persistent id · from dataverse fgv br
Catalogue records · 1
- Dataverse API dataverse.fgv.br/api/datasets/:persistentId/?persistentId=hdl%3A10438.3%2FFK2%2… ↗
metadata API · from dataverse fgv br
Topics
- Stated by source
- Computer and Information Science
- From keywords
- Computer Science & AI · Economics & Finance · Finance · Machine learning · Mathematics & Statistics
Provenance · 1 source records, 12 field assertions
| Source | Key | Last seen | Raw |
|---|---|---|---|
| FGV Dataverse | hdl:10438.3/FK2/PTGSZA | 18 h ago | JSON v1 |
| Field | Assertion | Extractor | Evidence |
|---|---|---|---|
| concepts[field].anzsrc:field:350202 | mapping · dataverse fgv br | vocabulary-mapper@1.0.0 | keywords['Finance'] |
| concepts[field].anzsrc:group:4611 | mapping · dataverse fgv br | vocabulary-mapper@1.0.0 | keywords['Machine learning'] |
| concepts[field].dataverse_subject:computer-and-information-science | source · dataverse fgv br | connector:dataverse_fgv_br@1.0.0 | /subjects |
| concepts[field].local:field:computer-science-ai | mapping · dataverse fgv br | connector:dataverse_fgv_br@1.0.0 | /subjects |
| concepts[field].local:field:economics-finance | mapping · dataverse fgv br | connector:dataverse_fgv_br@1.0.0 | /subjects |
| concepts[field].local:field:mathematics-statistics | mapping · dataverse fgv br | connector:dataverse_fgv_br@1.0.0 | /subjects |
| created_date | source · dataverse fgv br | connector:dataverse_fgv_br@1.0.0 | |
| description | source · dataverse fgv br | connector:dataverse_fgv_br@1.0.0 | /description |
| publication_date | source · dataverse fgv br | connector:dataverse_fgv_br@1.0.0 | |
| title | source · dataverse fgv br | connector:dataverse_fgv_br@1.0.0 | /name |
| updated_date | source · dataverse fgv br | connector:dataverse_fgv_br@1.0.0 | |
| version_label | source · dataverse fgv br | connector:dataverse_fgv_br@1.0.0 |