1. The 7 Reasons Most Machine Learning Funds Fail (Presentation Slides) by Marcos Lopez de Prado (Guggenheim Partners, LLC)
2. Craftsmanship Alpha: An Application to Style Investing by Ronen Israel, Sarah Jiang and Adrienne Ross (AQR Capital Management, LLC)
This paper stems from the growing popularity of style investing, and the general misconception that “all style portfolios are the same.” Style investing – also known as factor investing or smart beta – refers to taking advantage of proven sources of returns, such as value (cheap beats expensive), momentum (improving beats deteriorating) and others. Even though there is generally broad agreement on the major styles that drive investment returns, there are a variety of choices that can be made when constructing portfolios.
We use the term “craftsmanship alpha” to describe the skillful execution of style portfolios. When people talk about ‘alpha’ they generally mean trying to find new ways to generate investment returns. But there is a tremendous amount of alpha that you can extract just by focusing on better implementation of a strategy. We hope this paper helps investors better differentiate across style portfolios by understanding the nuances of implementation. – Adrienne Ross
3. Social Animal House: The Economic and Academic Consequences of Fraternity Membership by Jack Mara (10 Thoughts), Lewis Davis (Union College – Department of Economics) and Stephen Schmidt (Union College – Department of Economics)
4. Customer-Based Corporate Valuation for Publicly Traded Non-Contractual Firms by Daniel McCarthy (Emory University – Department of Marketing) and Peter Fader (University of Pennsylvania – Marketing Department)
In our previous paper, we laid out a methodology for customer-based corporate valuation for subscription firms, using a public company’s publicly disclosed customer data to more accurately value the firm and study its unit economics. While it was well received, it begged the question — is there an analogous way to value non-subscription businesses (such as e-commerce retailers) using their customer data? In the current paper, we answer this question. We propose a model for how customers are acquired, how long they remain with the firm, how many purchases they make while they are retained, and how much they spend on those purchases. These models are combined to generate the revenue forecasts which drive our valuation model. To bring the methodology to life, we apply it to data from two publicly traded e-commerce retailers, Overstock and Wayfair. We see this as an important step towards properly incorporating customer behaviors into standard valuation models. – Daniel McCarthy
5. The Decline of Violent Conflicts: What Do the Data Really Say? by Pasquale Cirillo (Delft University of Technology) and Nassim Taleb (NYU-Tandon School of Engineering)