Tracking ability of metal exchange traded funds (etfs)

• Main Author: Sousa, João Miguel Lança Tavares de
• Publication Date: 2014
• Format: Master thesis
• Language: eng
• Source: Repositórios Científicos de Acesso Aberto de Portugal (RCAAP)
• Download full: http://hdl.handle.net/10071/10341
• topic: Exchange traded funds; Commodities; Tracking ability; Tracking Error; Fundos de investimento

Exchange Traded Funds (ETFs) are investment funds traded in stock exchanges that have recently been introduced and have been growing significantly over the last few years. Despite the vast number of research studies on this topic, there has been a certain lack of attention directed to Commodity ETFs and their suitability as an alternative investment in commodities. In order to understand their viability as an investment and their capacity in tracking their benchmarks, this study analyzes a sample of 27 metal ETFs, traded on the NYSE Arca. The focus of this thesis is the weekly tracking ability of ETFs, evaluated through a linear regression, tracking error tests and performance analysis indicators. Then, it is also tested the potential mispricing to understand if the ETFs are traded at their “fair prices” or if their price tend to deviate from their net asset values (NAV). The linear regression and the tracking error results show that most ETFs have negative alphas, although not statistically significant. In addition, they also show strong correlations and low deviations from the benchmarks, where the physically-backed have the smallest values. Regarding the performance, it is not possible to elect the better strategy, due to the mixed results obtained. According to the mispricing analysis, most of the ETFs are traded at premium, where all the physically-backed are traded with a price higher than their NAV.