Investors take their financial decisions based on the relevant information that they attain about different assets, such as stocks and bonds. Since the advent of finance, as a field of money and investment, certain datasets – such as the balance sheet information or the historical returns on asset prices – have been widely used in both academic research as well as practical investment. Due to the ease of access, such datasets have been very popular by both academics and practitioners. At the same time, the popularity of such datasets reduces investors’ expected alpha for developing any strategy that only relies on such traditional datasets. Therefore, pioneering investors are looking for persistent sources of alpha in alternative datasets.
An alternative dataset can come from different sources such as social networks, mobile devices, news, satellites, public records, and the scraping of the internet.
Because of lack of competition in using such alternative datasets, the expected alpha is usually larger. Furthermore, due to its low correlation with the alpha that is attained from the traditional datasets, the alternative alpha can reduce portfolio risk, uncover opportunities, and improve portfolio performance. However, due to the unstructured and non-numerical nature and its vast volume, alternative datasets are often much more difficult to process than traditional datasets. The objective of this internship is to investigate a few alternative data sources and examine the possibility of attaining robust and persistent alpha. Therefore, at the quantitative research department of Robeco, we are looking for exceptional candidates with strong programming background to join us in this pioneering mission.
Karagozoglu and Fabozzi (2017) “Volatility wisdom of social media crowds”, Journal of Portfolio Management 43(2).
Katona, Painter, Patatoukas, and Zeng (2018) “On the capital market consequences of alternative data: evidence from outer space”, working paper.
Zhu (2019) “Big data as a governance mechanism”, Review of Financial Studies 32 (5)