Not logged in.
Quick Search - Contribution
Contribution Details
Type | Bachelor's Thesis |
Scope | Discipline-based scholarship |
Title | Influence of Futures-Positioning on Bond Returns: An ARIMAX forecasting model on COT data |
Organization Unit | |
Authors |
|
Supervisors |
|
Language |
|
Institution | University of Zurich |
Faculty | Faculty of Business, Economics and Informatics |
Number of Pages | 61 |
Date | 2022 |
Abstract Text | The research conducted in this thesis aims to investigate whether positioning data contained in the CFTC report of the two main investor groups, institutionals and retailers, reflects predictive power for future returns in the U.S. financial securities market. For tracking the positions and price evolution over di↵erent time horizons, multiple ARIMAX forecasting models were built. Those models were tested for statistical significancy and then implemented with the price as exogeneous variable and the long and short positions of the two investor groups as dependent variables, resulting in one-year forecasting plots. Even though some patterns are spotted, these one-factor models are not able to efficiently forecast future price movements, mainly due to the time lag between the publication date and the collecting of the data of the CFTC report, and the aggregate composition of the COT data. A possible way to solve these issues is to build a multi-factor model, as done by “Micaletti (2018)” in his paper “Want Smart Beta? Follow the Smart Money: Market and Factor Timing Using Relative Sentiment”. The study of this thesis adds to the current literature more empirical evidence on the statistical significance of trading using institutionals positioning data, confirming that, by its own, the forecasting power is relatively low. |
PDF File | Download |
Export | BibTeX |