Call For Papers

ADIA Lab Award for Causal Research in Investments

Hear from the Finalists!

After a successful call for papers in 2023, where we received more than 30 submissions, we have identified our top 5 finalists.

This series of seminars aims to showcase innovative research in causal research in investments and promote the use of the formal language of causal inference to express findings in finance and economics, with a particular focus on investing.

These presentations are part of the ADIA Lab "Best Paper Award" series for Causal Research in Investments, which seeks to elevate the formal language of causal inference in expressing findings in finance and economics, focusing on investing.

Motivation

Most journal articles in the investment literature make associational claims, and propose investment strategies designed to profit from those associations. For instance, authors may find that observation X often precedes the occurrence of event Y, determine that the correlation between X and Y is statistically significant, and propose a trading rule that presumably monetizes such correlation. A caveat of this inductive reasoning is that the probabilistic statement “X often precedes Y” provides no evidence that Y is a function of X, thus the relationship between X and Y may be coincidental or non-causal. The relationship is coincidental when variables X and Y appear to have been associated in the past, however that association is a statistical fluke, e.g. due to backtest overfitting. Backtest overfitting leads to false investment strategies that perform poorly out-of-sample. The relationship is non-causal when X and Y are associated even though Y is not a function of X, e.g. due to a confounding variable Z which researchers have failed to control for. Under-controlling makes it likely that the correlation between X and Y will change over time, and even reverse sign, exposing the investor to systematic losses.

Science is more than a collection of observed associations. While the description and cataloguing of phenomena play a role in scientific discovery, the ultimate goal of science is the amalgamation of theories that have survived rigorous falsification. For a theory to be scientific, it must declare the falsifiable causal mechanism responsible for the observed phenomenon. Knowledge of the causal mechanism in investments allows researchers to hedge unrewarded risks, decommission fading investment strategies before accumulating losses, and anticipate the possibility of black swans. Backtests and regressions cannot explain why a strategy may be successful. Only a falsifiable causal theory can give investors some confidence that the claimed performance is not based on spurious patterns and it is hence replicable going forward.

Scope

This call for papers aims to promote the use of the formal language of causal inference to express findings in the areas of finance and economics, with particular interest to investing. We welcome submissions that: (a) apply causal discovery algorithms to financial datasets; (b) use probabilistic graphical models to declare assumptions and assist in understanding the causal content of the equations; (c) identify confounders, mediators and colliders, explaining how the study uses them to block non-causal paths of association; (d) propose causal mechanisms that explain observed associations among financial variables; (e) propose misspecification tests, (actual or Gedanken) controlled experiments, natural experiments, and testable implications of the hypothesized causal mechanisms; and (f) present empirical evidence consistent with the causal theory, without engaging in backtest overfitting. We welcome using interpretable machine learning methods and novel alternative datasets to support causal claims. Authors may suggest a heuristic that monetizes the identified pattern, and report the backtested performance of that heuristic, however such evidence will be deemed inconclusive or anecdotal without a formal causal theory that explains the monetized pattern.

For example, a paper may hypothesize that, as some stocks achieve high valuations, large asset managers gradually reduce their exposure to those stocks. They may design natural experiments to back that claim, as explained in Lopez de Prado [2022]. In presenting their claims, researchers should avoid the pitfalls described in that publication.

More generally, the scope of this call is not limited to papers about investment strategies, and it encompasses contributions to the modeling of economic variables with investment implications. For example, papers may propose causal graphs to model the effects of various monetary policies, use Bayesian networks to implement a coherent stress testing framework, propose randomized controlled trials designed to assess the efficacy of broker algo-wheels, etc.

Prize

A total prize of USD 100,000 will be divided among the authors of the top three papers. The other two finalists will receive a diploma. Authors may publish their work in a journal of their choice.

Terms And Conditions

For a full description of the terms and conditions, please visit Terms and Conditions for Submissions. Anyone submitting a paper will be deemed to have read and agreed to these terms and conditions.