Exploring the value of alternative data and media sentiment in the investment process.
This white paper addresses the question of the value of alternative data in the investment process. We propose a quantitative framework to assess the added value of an alternative data set on the basis of a backtesting process, in combination with the so-called GH1 measure, which takes into account both return enhancement and risk reduction, with respect to a particular benchmark.
“We demonstrate that this alternative data set provides significant value to investors: using media sentiment as the single factor achieves the same investment results as the full-blown multifactor strategy.”
"Using the GH1 measure, one can determine whether the combination of the index and the risk free asset generates higher return at the same level of risk than the fund that uses alternative data."
"We base our portfolio construction on the so-called alpha-momentum strategy by Hühn and Scholz (2018). This strategy ranks stocks based on their alphas from the factor model."
"The data set we use as the example of an alternative data source is the news sentiment data from LSEG News Analytics, based on complex natural language processing (NLP) algorithms, which ‘reads’ and interprets (in real time) all news that reaches the Reuters news wire."
This research provides an illuminating framework for assessing the added value of alternative data with respect to any benchmark, be it a passive index or an active, factor-based investment strategy.