University of Illinois at Urbana-Champaign
Can intraday data improve commodity hedging performance?
Abstract
dc:descriptionWe examine if there is economic value in using intraday data to hedge commodity spot prices in the futures market using the realized minimum-variance hedging ratio (RMVHR) framework. The latter depends on the forecast of the realized futures-cash covariance matrix. We consider both multiple and single commodity portfolios that relate to the crude oil crack and soybean crush industries, as well as different forecast strategies. Forecasts are built based on the heterogeneous autoregressive (HAR) model and range from a direct forecast of the RMVHR, a forecast of the futures-spot covariance for a single commodity to then build the RMVHR, to a multi-commodity portfolio where inter-commodity spillovers are also forecast. We use the Naïve hedging ratio as the benchmark to investigate the performance of intraday data-based hedging models. Our results suggest that for each portfolio considered, there is at least one intraday data-based hedging strategy that outperforms the Naïve in the soybean crush complex. In the crude oil crush complex, the superiority of using intraday data is not always statistically significant. Forecasting RMVHR directly is generally considered the best strategy, which suggests that simpler is better. Our estimates place the advantage of using intraday data between USD7,305 and USD250 per contract and year on average, with these values representing the decline in the portfolio’s standard deviation achieved through hedging.
Degree
thesis:*- Name thesis:degree_name
- M.S.
- Level thesis:degree_level
- Thesis
- Discipline thesis:degree_discipline
- Agricultural & Applied Econ
- Grantor
- University of Illinois at Urbana-Champaign
- Year dc:date
- 2022
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Wu, Shujie
- Contributors dc:contributor
-
- Serra Devesa, Maria Teresa
- Garcia, Phillip
- Irwin, Scott
Subjects
dc:subject × 5Rights
dc:rights- Statement dc:rights
-
- Copyright 2022 Shujie Wu
- Language dc:language
- en, eng
Identifiers
dc:identifier.*- Handle dc:identifier
- https://hdl.handle.net/2142/117665