Call For Papers

ADIA Lab Award for Causal Research in Investments

Hear from the Finalists!


Five finalists invited to present their papers in one of our “ADIA Lab Best Paper Award” calls.

After our successful call for papers in 2023, where we received more than 30 submissions, we have identified our top 5 finalists, and we are excited to invite you to join us for their virtual presentations over the coming weeks. Below, you will find the dates for all five seminars as well as the links to the papers. Please sign up to receive the calendar invites, including the MS Teams links. Whether you choose to attend all five seminars or just the ones that pique your interest the most, we welcome your participation. This series of seminars aims to showcase innovative research in the field of 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.

The presentations will take place on the following dates, all at 7PM GST:

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

Motivation

The majority of 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 plays 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. In the context of investments, knowledge of the causal mechanism allows researchers to hedge unrewarded risks, decommission fading investment strategies before they accumulate 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 the use of 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 in absence of 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.

Timeline

The review of papers will proceed according to the following timeline:

  • Papers must be submitted before the deadline, set to December 31, 2023 at midnight (Abu Dhabi time)

    • Submitted papers must be available in a public repository, such as SSRN, ArXiv, or a journal website

    • Valid submissions include:

      • unpublished papers

      • papers published in pre-print form or under journal review

      • journal articles published after January 1, 2022. Papers available behind a paywall are eligible

    • To submit a publication, authors must send an e-mail to the address Paper.Award.2023@adialab.ae, including:

      • Cover letter

      • Link where the publication can be accessed. Authors should not send the paper as an attachment to the e-mail

      • Names, affiliations and contact details of author.

  • Following a blind peer-review, in April 2024 the organizers will

    • select up to 5 finalists.

    • notify the authors.

    • announce the selected articles in ADIA Lab’s website, including the links to the public repositories where the articles are stored. Unselected submissions will remain confidential.

  • In April and May 2024, the finalists will be invited to give public online presentations, with one presentation per week, over a time of five week.

  • In June 2024, the winners will be announced in a public event.

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.

Prize

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.