This study presents evidence that much of the trading on macro-economic news occurs prior to the scheduled news announcement times. Examining the trading patterns ahead of the ISM Manufacturing Index and Construction Spending announcement, we find that the trading on the not-yet-publicly...

This article compares the performance of systematic and discretionary hedge funds. In systematic funds, most trades are originated and executed by computer programs written for that purpose. In discretionary funds, human traders make decisions when to buy and sell which financial securities. The...

We derive the economic costs of latency induced by computer technology in trading. We show that the costs of latency are negligible in their expected value, but instead manifest themselves in increased risk to investors. We also show that our theoretical predictions firmly hold when tested on...

Using Artificial Intelligence (AI) techniques, we quantitatively examine the mentions of ESG terms in the U.S. corporate filings with the SEC over the 2019-20 period. We find that in our sample of companies, the management tends to focus ESG discussions in corporate filings along three...

We show a simple way to let the data speak for itself. Specifically, we show how a large mixed bag of data, potentially embedded with missing data points and collinearities, and therefore unsuitable for traditional econometric analysis, can be useful in building fast and meaningful big data and...

The current research assesses the risks commonly attributed to the presence of HFT in the context of different market structures deployed by the U.S. exchanges. In particular, we find that, by design, the so-called “normal” exchanges have the lowest market quality, including the highest...

Corporate disclosures have been studied for many years. We add a new dimension to the analysis: we show that the filing frequency of disclosures has a negative impact on future stock returns. Specifically, we show that every additional filing within the past month preceding a given filing...

Six key problems are known to plague traditional econometric analysis and much of pure machine learning: 1) overfitting, 2) inability to deal with noise in the data, 3) difficulties dealing with missing observations, 4) data non-stationarity, 5) researcher ability and creativity, and 6)...

The crypto market is growing at an unprecedented rate, but it can be difficult to make money in this market if you don't understand how it works. Most people are still using traditional methods to trade in the crypto market, which can lead to losses due to significantly adverse pricing, known as...

We show a simple way to let the data speak for themselves. Specifically, we show how a large mixed bag of data, potentially embedded with missing data points and collinearities, and therefore unsuitable for traditional econometric analysis, can be useful in building fast and meaningful big data...