Beating the Market
Charting by machines
Scott Murray, Yusen Xia & Houping Xiao
Journal of Financial Economics, March 2024
Abstract:
We test the efficient market hypothesis by using machine learning to forecast stock returns from historical performance. These forecasts strongly predict the cross-section of future stock returns. The predictive power holds in most subperiods and is strong among the largest 500 stocks. The forecasting function has important nonlinearities and interactions, is remarkably stable through time, and captures effects distinct from momentum, reversal, and extant technical signals. These findings question the efficient market hypothesis and indicate that technical analysis and charting have merit. We also demonstrate that machine learning models that perform well in optimization continue to perform well out-of-sample.
Megatrends and the U.S. Economy, 1890-2040
Joseph Davis & Lukas Brandl-Cheng
Vanguard Group Working Paper, January 2024
Abstract:
Will we have a future of too-few jobs due to AI or too-few workers due to a demographic drag? We estimate the potential impact that five megatrends—technology, demographics, fiscal deficits, globalization, and energy transitions—may have on future U.S. economic and financial outcomes. In doing so, this paper breaks new ground along several dimensions. First, we develop a new quarterly dataset that begins in 1890 to capture historical shifts in technology and demographics. Second, by casting the supply-side of the economy in time-varying trend growth rates, we can jointly model “supply-side” megatrends with “demand-side” cyclical variables such as output gap and inflation in a large Bayesian VAR framework with multivariate stochastic volatility. This approach finds that megatrend shocks have accounted for nearly 50% of the cyclical variance in real GDP and the stock market over the past 130 years. We develop a novel identification strategy using a combination of zero and sign restrictions that is grounded in key insights from both task-based endogenous growth and cyclical New-Keynesian models. We uniquely identify eleven structural shocks, including three types of technology shocks (general purpose technology, automation, and labor-augmenting). We decompose the nominal policy rate to identify real monetary policy shocks, inflation expectational shocks, and structural fiscal deficit shocks (excluding wars and recessions) that allow us to endogenously date regimes of “active” and “passive” monetary policy. Our historical decompositions refine or challenge some stylized facts regarding the relative importance of megatrends and policy for business cycles, interest rates, inflation, and productivity growth. Finally, our structural megatrend scenarios through 2040 reveal future growth and inflation distributions that are bimodal, reflecting an emerging horserace between a “Productivity Surges” scenario, where AI’s productivity impact through automation exceeds that of the computer and Internet and raises trend growth, and a “Demographics Dominate” scenario, where structural fiscal deficits worsen due to an aging society that raise real borrowing costs.
Judge Ideology and Opportunistic Insider Trading
Allen Huang, Kai Wai Hui & Yue Zheng
Journal of Financial and Quantitative Analysis, forthcoming
Abstract:
Although federal judges are the ultimate arbiters of insider trading enforcement, the role of their political ideology in insider trading is unclear. Using the partisanship of judges’ nominating presidents to measure judge ideology, we first document that liberal judges are associated with heavier penalties in insider trading lawsuits than conservative judges. Next, we find that firms located in circuits with more liberal judges have fewer opportunistic insider sales. Cross-sectional analyses show that this deterrent effect is stronger when managers face a higher risk of insider trading lawsuits. Finally, we find that the SEC considers judges’ ideology when selecting litigation forums.
A Cognitive Foundation for Perceiving Uncertainty
Aislinn Bohren et al.
NBER Working Paper, February 2024
Abstract:
We propose a framework where perceptions of uncertainty are driven by the interaction between cognitive constraints and the way that people learn about it -- whether information is presented sequentially or simultaneously. People can learn about uncertainty by observing the distribution of outcomes all at once (e.g., seeing a stock return distribution) or sampling outcomes from the relevant distribution sequentially (e.g., experiencing a series of stock returns). Limited attention leads to the overweighting of unlikely but salient events -- the dominant force when learning from simultaneous information -- whereas imperfect recall leads to the underweighting of such events—the dominant force when learning sequentially. A series of studies show that, when learning from simultaneous information, people are overoptimistic about and are attracted to assets that mostly underperform, but sporadically exhibit large outperformance. However, they overwhelmingly select more consistently outperforming assets when learning the same information sequentially, and this is reflected in beliefs. The entire 40-percentage point preference reversal appears to be driven by limited attention and memory; manipulating these factors completely eliminates the effect of the learning environment on choices and beliefs, and can even reverse it.
Art in times of crisis
Géraldine David et al.
Economic History Review, forthcoming
Abstract:
We trace the long-term performance of the UK art market across a broad set of crises: world wars, economic recessions, financial crises, inflationary periods, and changes in monetary policy. By means of digitalized historical auction archives, we construct art price indices from the early twentieth century onwards and disclose that annual art auction value grew, in real terms, more than seven-fold over this period. The arithmetic annual real return and risk amount to 3.6 per cent and 20.1 per cent, respectively. Art returns plummeted at the onset of wars, but turned positive in the second half of wars when they outperformed stocks, suggesting that art was seen as a safe haven in times of political turmoil. During wars, smaller -- and thus more transportable -- paintings obtained higher returns. Art returns are sensitive to economic and financial crises, with the largest slumps occurring during the Great Depression, oil crisis, recessions of the early 1980s and early 1990s, and the Great Recession. We also document changes in art preferences for paintings’ sizes, schools, liquid art, and artists’ nationalities across crises. Art enters a broad optimal asset portfolio both in non-crisis periods and during war times.
Paying Managers of Complex Portfolios: Evidence on Compensation and Performance from Endowments
Matteo Binfarè & Robert Harris
Journal of Financial and Quantitative Analysis, forthcoming
Abstract:
We examine compensation for endowment Chief Investment Officers (CIOs) overseeing portfolios with significant allocations to alternatives. We find widespread use of bonuses and that large endowments with high alternative allocations hire CIOs with stronger backgrounds, pay them more, and have higher pay-for-performance sensitivity. We find weak evidence of a relationship between compensation and future performance. Our results align with contract theory predictions but differ from empirical findings on pension funds. Endowments pay CIOs more, rely more on bonuses, attract more experienced professionals, and have lower turnover than pensions. This suggests more effective talent management compared to politically influenced public pensions.
New News is Bad News
Paul Glasserman, Harry Mamaysky & Jimmy Qin
Columbia University Working Paper, August 2023
Abstract:
An increase in the novelty of news predicts negative stock market returns and negative macroeconomic outcomes over the next year. We quantify news novelty -- changes in the distribution of news text -- through an entropy measure, calculated using a recurrent neural network applied to a large news corpus. Entropy is a better out-of-sample predictor of market returns than a collection of standard measures. Cross-sectional entropy exposure carries a negative risk premium, suggesting that assets that positively covary with entropy hedge the aggregate risk associated with shifting news language. Entropy risk cannot be explained by existing long-short factors.
Inefficient Forecasts at the Sportsbook: An Analysis of Real-Time Betting Line Movement
Jay Simon
Management Science, forthcoming
Abstract:
This paper tests the efficiency of a set of sports betting markets using detailed betting line movement from opening until closing for four different sportsbooks for each of 3,681 Major League Baseball games. The reliability of the markets’ forecasts are assessed at several lead times. The forecasts are mostly reliable, but there are simple betting strategies that would have yielded significant profit. In addition, forecasts do not always improve monotonically as the games get closer, despite more information being available; forecasts at weekend day games’ start times are significantly worse than forecasts 90 minutes earlier. Furthermore, analysis of the sequences of forecasts within individual games reveals that these betting markets do not incorporate information optimally. There is sufficient evidence to reject weak form market efficiency; specifically, betting lines tend to overreact, exhibiting significant negatively autocorrelated changes that could be exploited by sophisticated bettors.