Refine
Document Type
- Working Paper (3)
Language
- English (3)
Has Fulltext
- yes (3)
Is part of the Bibliography
- no (3)
Keywords
- Liquidity (2)
- Algorithmic Trading (1)
- Automation (1)
- Intermediary (1)
- Inventory Risk (1)
- Microstructure (1)
- Volatility (1)
- Wertpapierhandelssystem (1)
Algorithmic trading has sharply increased over the past decade. Equity market liquidity has improved as well. Are the two trends related? For a recent five-year panel of New York Stock Exchange (NYSE) stocks, we use a normalized measure of electronic message traffic (order submissions, cancellations, and executions) as a proxy for algorithmic trading, and we trace the associations between liquidity and message traffic. Based on within-stock variation, we find that algorithmic trading and liquidity are positively related. To sort out causality, we use the start of autoquoting on the NYSE as an exogenous instrument for algorithmic trading. Previously, specialists were responsible for manually disseminating the inside quote. As stocks were phased in gradually during early 2003, the manual quote was replaced by a new automated quote whenever there was a change to the NYSE limit order book. This market structure change provides quicker feedback to traders and algorithms and results in more message traffic. For large-cap stocks in particular, quoted and effective spreads narrow under autoquote and adverse selection declines, indicating that algorithmic trading does causally improve liquidity.
Price pressures
(2010)
We study price pressures in stock prices—price deviations from fundamental value due to a risk-averse intermediary supplying liquidity to asynchronously arriving investors. Empirically, twelve years of daily New York Stock Exchange intermediary data reveal economically large price pressures. A $100,000 inventory shock causes an average price pressure of 0.28% with a half-life of 0.92 days. Price pressure causes average transitory volatility in daily stock returns of 0.49%. Price pressure effects are substantially larger with longer durations in smaller stocks. Theoretically, in a simple dynamic inventory model the ‘representative’ intermediary uses price pressure to control risk through inventory mean reversion. She trades off the revenue loss due to price pressure against the price risk associated with remaining in a nonzero inventory state. The model’s closed-form solution identifies the intermediary’s relative risk aversion and the distribution of investors’ private values for trading from the observed time series patterns. These allow us to estimate the social costs—deviations from constrained Pareto efficiency—due to price pressure which average 0.35 basis points of the value traded. JEL Classification: G12, G14, D53, D61
Non-standard errors
(2021)
In statistics, samples are drawn from a population in a data-generating process (DGP). Standard errors measure the uncertainty in sample estimates of population parameters. In science, evidence is generated to test hypotheses in an evidence-generating process (EGP). We claim that EGP variation across researchers adds uncertainty: non-standard errors. To study them, we let 164 teams test six hypotheses on the same sample. We find that non-standard errors are sizeable, on par with standard errors. Their size (i) co-varies only weakly with team merits, reproducibility, or peer rating, (ii) declines significantly after peer-feedback, and (iii) is underestimated by participants.
