Roger M. Stein

Funding Long Shots

A project with a low probability of success can be difficult to fund even if the expected return is high and the correlation with other investments is low. This paper describes how such projects can be combined in a fund, and securitization techniques used, so that they are attractive to investors. It examines how the proposed structure is affected by the number of projects in the fund, the probability of each project’s success, the projects’ life, correlations between the success of different projects, and uncertainty about the eventual payoff from a project if it is successful.

Systemic risk and alternative investments:A summary of selections from the state of the art

Alternative investments, by definition, are chosen by investors to be relatively less correlated with broad market factors. However, recent events have demonstrated that during periods of extreme market dislocation, even alternative assets may be affected by market dysfunction. In this survey, which serves as an introduction to this Special Issue, the editors summarize a number of articles from some of the leading researchers currently studying systemic risk to provide readers with a sampling of some of the contours of the landscape of this important field. Although far from comprehensive, the perspectives of these articles give a sense of the wide range of issues involved in measuring systemic risk and the corresponding diversity of methods for understanding them—often bringing together experts and methods from outside of traditional finance. Importantly, there is no consensus regarding a single measure or method for understanding systemic risk. In general, most researchers in both academia and the industry favor using a number of approaches, depending on the questions being explored; thus it benefits readers to get a broad view of the various emerging approaches. While the editors have tried to provide one such view, they note that the snapshot these articles present is not intended as a primer on systemic risk. Rather, they provide an enticing in vivo sampling of where this dynamic and expanding field is and where it appears to be headed.

FinTech Platforms and Strategy

A major consequence of the Internet era is the emergence of complex “platforms�? that combine technology and process in new ways that often disrupt existing industry structures and blur industry boundaries. These platforms allow easy participation that often strengthens and extends network effects, while the vast amounts of data captured through such participation can increase the value of the platform to its participants, creating a virtuous cycle. While initially slow to penetrate the financial services sector, such platforms are now beginning to emerge. We provide a taxonomy of platforms in Finance and identify the feasible strategies that are available to incumbents in the industry, innovators, and the major Internet giants.

Evaluating discrete choice prediction models when the evaluation data is corrupted: analytic results and bias corrections for the area under the ROC

There has been a growing recognition that issues of data quality, which are routine in practice, can materially affect the assessment of learned model performance. In this paper, we develop some analytic results that are useful in sizing the biases associated with tests of discriminatory model power when these are performed using corrupt (“noisy”) data. As it is sometimes unavoidable to test models with data that are known to be corrupt, we also provide some guidance on interpreting results of such tests. In some cases, with appropriate knowledge of the corruption mechanism, the true values of the performance statistics such as the area under the ROC curve may be recovered (in expectation), even when the underlying data have been corrupted. We also provide estimators of the standard errors of such recovered performance statistics. An analysis of the estimators reveals interesting behavior including the observation that “noisy” data does not “cancel out” across models even when the same corrupt data set is used to test multiple candidate models. Because our results are analytic, they may be applied in a broad range of settings and this can be done without the need for simulation.