Please use this identifier to cite or link to this item:
https://doi.org/10.21256/zhaw-29317| Publication type: | Master thesis |
| Title: | Hedge fund replication through trend-following |
| Authors: | Gjergji, Zef |
| Advisors / Reviewers: | Boos, Dominik |
| DOI: | 10.21256/zhaw-29317 |
| Extent: | 76 |
| Issue Date: | 2022 |
| Publisher / Ed. Institution: | ZHAW Zürcher Hochschule für Angewandte Wissenschaften |
| Publisher / Ed. Institution: | Winterthur |
| Language: | English |
| Subject (DDC): | 332.6: Investment |
| Abstract: | While the use of managed futures funds has increased in the past, the actual investment strategy remains to some extent a secret to outside investors. In recent years, managed futures have often been associated with trend-following. Indeed, several studies have shown that trend-following strategies can significantly replicate the returns of these funds. However, it is not known exactly what type of trend-following strategies are used in practice. The current state of the literature shows that time series momentum, which only considers the past return of the underlying asset itself, largely explains the returns of managed futures. However, recent academic evidence has shown that the weighting coefficients of speculators' returns in futures markets do not resemble the weightings of time series momentum strategies. They appear to be more similar to the more dynamic moving average crossover strategies. On this basis, it was hypothesized that moving average crossover signals can better replicate the returns of managed futures than time series momentum. Related to this is the research question of which trend-following signals most clearly explain managed futures returns. To answer this research question, three widely adopted trend-following strategies were first developed and analyzed. These strategies consisted of time series momentum and different moving average crossovers. They were then categorized into short-, medium-, and long-term trend signals. In terms of underlying data, the focus was on liquid futures contracts and currency pairs. Comparing all signals with conventional passive investments in different markets and common risk factors, there is a positive significant alpha left for each signal and each considered trend horizon. This result is evidence of the presence and importance of momentum for the three different trend following strategies and also shows low exposure to common investment factors. In separate regressions in this thesis, moving average crossover strategies were found to be more capable of explaining managed futures returns than time series momentum strategies. A subsequent analysis of different approaches to risk assessment also revealed that managed futures incorporate a type of widely used RiskMetrics approach that has not been previously discussed in the literature. Based on the previous scientific finding that the distinction between long- and short-only trend signals can better replicate the returns of managed futures, this separation was also analyzed. However, since the statistical regressions were strongly affected by multicollinearity between these split signals, a relaxed lasso procedure was applied to obtain a representative and stable set of trend-following signals that can accurately describe the returns of managed futures. The derived signal selection showed promising explanatory values and furthermore novel robust values in the predictability of managed futures returns. Furthermore, it was possible to identify certain investment preferences of managed futures. |
| URI: | https://digitalcollection.zhaw.ch/handle/11475/29317 |
| License (according to publishing contract): | CC BY-NC-ND 4.0: Attribution - Non commercial - No derivatives 4.0 International |
| Departement: | School of Management and Law |
| Appears in collections: | MSc Banking and Finance |
Files in This Item:
| File | Description | Size | Format | |
|---|---|---|---|---|
| 2022_Zef_Gjergji_MSc_BF.pdf | 1.43 MB | Adobe PDF |  View/Open |
Show full item record
Gjergji, Z. (2022). Hedge fund replication through trend-following [Master’s thesis, ZHAW Zürcher Hochschule für Angewandte Wissenschaften]. https://doi.org/10.21256/zhaw-29317
Gjergji, Z. (2022) Hedge fund replication through trend-following. Master’s thesis. ZHAW Zürcher Hochschule für Angewandte Wissenschaften. Available at: https://doi.org/10.21256/zhaw-29317.
Z. Gjergji, “Hedge fund replication through trend-following,” Master’s thesis, ZHAW Zürcher Hochschule für Angewandte Wissenschaften, Winterthur, 2022. doi: 10.21256/zhaw-29317.
GJERGJI, Zef, 2022. Hedge fund replication through trend-following. Master’s thesis. Winterthur: ZHAW Zürcher Hochschule für Angewandte Wissenschaften
Gjergji, Zef. 2022. “Hedge Fund Replication through Trend-Following.” Master’s thesis, Winterthur: ZHAW Zürcher Hochschule für Angewandte Wissenschaften. https://doi.org/10.21256/zhaw-29317.
Gjergji, Zef. Hedge Fund Replication through Trend-Following. ZHAW Zürcher Hochschule für Angewandte Wissenschaften, 2022, https://doi.org/10.21256/zhaw-29317.
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.