A critical task in stock selection is identifying a firm’s true profitability. Given the potential of AI to deal with large data, an important question is: Can AI outsmart seasoned analysts?
If the task is to identify a firm’s true profitability, can AI outsmart seasoned analysts?
There’s no reason to think that the use of AI should lead to persistent fund outperformance, with any advantages gained likely being short lived.
AI-powered growth concentrates among larger firms and is associated with higher industry concentration. Our results highlight that new technologies like AI can contribute to growth and superstar firms through product innovation.
An AI analyst trained to digest corporate disclosures, industry trends, and macroeconomic indicators surpasses most analysts in stock return predictions. AI wins when information is transparent but voluminous. Humans provide significant incremental value in “Man + Machine,” which also substantially reduces extreme errors.
Without question the topic of greatest debate among investors, including investment professionals, and financial economists, is whether or not the market, and the technology sector in particular, is overvalued. There are two very strong conflicting views regarding not only the current valuation of technology stocks, but also the valuation of the entire asset class of large-cap growth stocks. One side, I’ll call the “new paradigm” or “it’s different this time” school. The other side, I’ll call “the been there, done that” school. Its theme is those that don’t learn from the past are doomed to repeat the same mistakes. No two sides could have more different viewpoints. To understand each side, let’s imagine a dialogue between the two schools.
Trading costs, discontinuous trading, missed trades, and other frictions, along with asset management fees can cause a shortfall between live and paper portfolios. The focus of this paper is to test an effective rebalancing method that prioritizes trades with the strongest signals to capture more of the factor premia while reducing turnover and trading costs.
Simple models severely understate return predictability compared to “complex” models in which the number of parameters exceeds the number of observations.
To date, the best metric we have for forecasting future equity returns and the ERP is current valuations. An interesting question is whether more complicated methods using newly developed machine learning models can provide superior forecasts.
Can AI models improve on the failures in predicting returns strictly from a practical point of view? In this paper, the possibilities are tested with a battery of AI models including linear regression, dimensional reduction methods, regression trees and neural networks. These machine learning models may be better equipped to address the multidimensional nature of stock returns when compared to traditional sorting and cross-sectional regressions used in factor research. The authors hope to overcome the drawbacks and confirm the results of traditional quant methods. As it turns out, those hopes are only weakly fulfilled by the MLM framework.
While there is literature that describes the "domain" of artificial intelligence, there are very few, if any that analyze the valuation and pricing of AI stocks. The authors attempt to fill the void with a two part methodology.
Along with the rapid growth in the utilization of robo-advisors, there has been similar growth in academic interest about robo-advisors. What is the current state and what are the main research streams in the literature?
One use of the NLP (natural language processing) features of ChatGPT is to search out patterns in the immense amounts of news, data and other sources of information about specific stocks, and then efficiently convert them into summaries valuable for all types of investors. Can this be accomplished with useful results? The authors use the Q2_2023 period to test performance around earnings announcements. Earnings announcements and earnings surprises are informationally rich as well as challenging events for investors to analyze.
Can machine-learning methods be used to predict the performance of active mutual funds, specifically in terms of alpha net of all costs? Answer: yes.
In this study, we introduce a novel measure of information content (Human-AI Differences, HAID) by exploiting the discrepancy between answers to questions at earnings calls provided by corporate executives and those given by several context-preserving Large Language Models (LLM) such as ChatGPT, Google Bard, and an open source LLM.
BloombergGPT is a large language model (LLM) developed specifically for financial tasks. The authors trained the LLM on a large body of financial textual data, evaluated it on several financial language processing tasks and found it performed at a significantly higher level than several other state-of-the-art LLMs.
Can chatbots, like ChatGPT, be used to interpret and condense lengthy financial disclosures into shorter but relevant documents?
This article examines the state of Artificial Intelligence (AI). We examine its history with an eye toward what it may mean for the world in years to come.
The paper documents that return forecasts from machine learning methods lead to superior out-of-sample returns in emerging markets.
We examine Sentix sentiment indices for use in tactical asset allocation. In particular, we construct monthly relative sentiment factors for the U.S., Europe, Japan, and Asia ex-Japan by taking the difference in 6-month economic expectations between each region's institutional and individual investors. These factors (along with one-month forward equity returns) then serve as inputs to a wide array of machine learning algorithms. Employing combinatorial cross-validation and adjusting for data snooping, we find relative sentiment factors have robust and significant predictive power in all four regions; that they surpass both standalone sentiment and time-series momentum in terms of informational content; and that they demonstrate the ability to identify the subsequent best- and worst-performing global equity markets from along a cross-section. The results are consistent with previous findings on relative sentiment, discovered using unrelated datasets.
© Copyright 2023 alpha architect | All Rights Reserved | Home | Terms of Use | Privacy Policy | Disclosures | Subscribe | Contact Us
