nep-mst New Economics Papers
on Market Microstructure
Issue of 2017‒11‒19
seven papers chosen by
Thanos Verousis
University of Newcastle

  1. Financial market predictions with Factorization Machines: Trading the opening hour based on overnight social media data By Stübinger, Johannes; Walter, Dominik; Knoll, Julian
  2. High Frequency Trading and Fragility By Giovanni Cespa; Xavier Vives
  3. Noise Traders Incarnate: Describing a Realistic Noise Trading Process By Peress, Joël; Schmidt, Daniel
  4. Measuring Price Discovery between Nearby and Deferred Contracts in Storable and Non-Storable Commodity Futures Markets By Zhepeng Hu; Mindy Mallory; Teresa Serra; Philip Garcia
  5. How Biased is the Behavior of the Individual Investor in Warrants? By Margarida Abreu
  6. Modelling and mitigation of Flash Crashes By Fry, John; Serbera, Jean-Philippe
  7. Long-range Auto-correlations in Limit Order Book Markets: Inter- and Cross-event Analysis By Martin Magris; Jiyeong Kim; Esa Rasanen; Juho Kanniainen

  1. By: Stübinger, Johannes; Walter, Dominik; Knoll, Julian
    Abstract: This paper develops a statistical arbitrage strategy based on overnight social media data and applies it to high-frequency data of the S&P 500 constituents from January 2014 to December 2015. The established trading framework predicts future financial markets using Factorization Machines, which represent a state-of-the-art algorithm coping with high-dimensional data in very sparse settings. Essentially, we implement and analyze the effectiveness of support vector machines (SVM), second-order Factorization Machines (SFM), third-order Factorization Machines (TFM), and adaptive-order FactorizationMachines (AFM). In the back-testing study, we prove the efficiency of Factorization Machines in general and show that increasing complexity of Factorization Machines provokes higher profitability - annualized returns after transaction costs vary between 5.96 percent for SVM and 13.52 percent for AFM, compared to 5.63 percent of a naive buy-and-hold strategy of the S&P 500 index. The corresponding Sharpe ratios range between 1.00 for SVM and 2.15 for AFM. Varying profitability during the opening minutes can be explained by the effects of market efficiency and trading turmoils. Additionally, the AFM approach achieves the highest accuracy rate and generates statistically and economically remarkable returns after transaction costs without loading on any systematic risk exposure.
    Keywords: finance,social media data,Factorization Machine,overnight information,statistical arbitrage,high-frequency trading
    Date: 2017
  2. By: Giovanni Cespa; Xavier Vives
    Abstract: We show that limited dealer participation in the market, coupled with an informational friction resulting from high frequency trading, can induce demand for liquidity to be upward sloping and strategic complementarities in traders’ liquidity consumption decisions: traders demand more liquidity when the market becomes less liquid, which in turn makes the market more illiquid, fostering the initial demand hike. This can generate market instability, where an initial dearth of liquidity degenerates into a liquidity rout (as in a flash crash). While in a transparent market, liquidity is increasing in the proportion of high frequency traders, in an opaque market strategic complementarities can make liquidity U-shaped in this proportion as well as in the degree of transparency.
    Keywords: market fragmentation, high frequency trading, flash crash, asymmetric information
    JEL: G10 G12 G14
    Date: 2016
  3. By: Peress, Joël; Schmidt, Daniel
    Abstract: We estimate a realistic process for noise trading to help theorists calibrate noisy rational expectations models. For this purpose, we characterize the trades initiated by individual investors, who are natural candidates for the role of noise traders because their trades are, on average, cross-correlated and loss making. We use transactions data from a retail brokerage house, small TAQ trades, and flows to retail mutual funds, obtaining consistent results. We find that noise trading can be treated as approximately i.i.d. at monthly and lower frequencies but that weekly and daily trades are serially correlated; the distribution of noise trading is less heavy-tailed at lower frequency but conforms to a normal only for quarterly data. We provide a complete description of these processes, including estimates of their standard deviation. In line with theory, the estimates are higher for more liquid and volatile stocks; they also suggest that the prevalence of noise trading has declined over time.
    Date: 2017–11
  4. By: Zhepeng Hu; Mindy Mallory; Teresa Serra; Philip Garcia
    Abstract: Futures market contracts with varying maturities are traded concurrently and the speed at which they process information is of value in understanding the pricing discovery process. Using price discovery measures, including Putnins (2013) information leadership share and intraday data, we quantify the proportional contribution of price discovery between nearby and deferred contracts in the corn and live cattle futures markets. Price discovery is more systematic in the corn than in the live cattle market. On average, nearby contracts lead all deferred contracts in price discovery in the corn market, but have a relatively less dominant role in the live cattle market. In both markets, the nearby contract loses dominance when its relative volume share dips below 50%, which occurs about 2-3 weeks before expiration in corn and 5-6 weeks before expiration in live cattle. Regression results indicate that the share of price discovery is most closely linked to trading volume but is also affected, to far less degree, by time to expiration, backwardation, USDA announcements and market crashes. The effects of these other factors vary between the markets which likely reflect the difference in storability as well as other market-related characteristics.
    Date: 2017–11
  5. By: Margarida Abreu
    Abstract: Based on the actual trading behavior of individual investors in the Portuguese financial market during almost ten years this paper examines the socio-demographic characteristics of retail investors in warrants, and discusses the hypothesis that some behavioral biases do have an impact on the investors’ predisposition to invest and trade in warrants, a complex financial instrument. One finds that there is a profile of investors in warrants: younger and less educated men are more likely to invest in warrants and that overconfident,disposition-prone and investors exhibiting a gambling attitude are more likely to invest and trade in warrants. Secondly, the gambling motive seems to be a distinguishing characteristic of investors in warrants. In other words, when investors are driven to trade in financial markets for pleasure/fun they tend to trade complex products more and to trade simple and easier to understand financial instruments less. Finally, the higher the intensity of trading the more relevant are the disposition and the gambler’s biases.
    Keywords: Warrants, overconfidence, disposition effect, gambling effect, individual investor behavior
    JEL: G11 G12
    Date: 2017–10
  6. By: Fry, John; Serbera, Jean-Philippe
    Abstract: The algorithmic trading revolution has had a dramatic effect upon markets. Trading has become faster, and in some ways more efficient, though potentially at the cost higher volatility and increased uncertainty. Stories of predatory trading and flash crashes constitute a new financial reality. Worryingly, highly capitalised stocks may be particularly vulnerable to flash crashes. Amid fears of high-risk technology failures in the global financial system we develop a model for flash crashes. Though associated with extreme forms of illiquidity and market concentration flash crashes appear to be unpredictable in advance. Several measures may mitigate flash crash risk such as reducing the market impact of individual trades and limiting the profitability of high-frequency and predatory trading strategies.
    Keywords: Flash Crashes; Flash Rallies; Econophysics; Regulation
    JEL: C1 F3 G1 K2
    Date: 2017–09–12
  7. By: Martin Magris; Jiyeong Kim; Esa Rasanen; Juho Kanniainen
    Abstract: Long-range correlation in financial time series reflects the complex dynamics of the stock markets driven by algorithms and human decisions. Our analysis exploits ultra-high frequency order book data from NASDAQ Nordic over a period of three years to numerically estimate the power-law scaling exponents using detrended fluctuation analysis (DFA). We address inter-event durations (order to order, trade to trade, cancel to cancel) as well as cross-event durations (time from order submission to its trade or cancel). We find strong evidence of long-range correlation, which is consistent across different stocks and variables. However, given the crossovers in the DFA fluctuation functions, our results indicate that the long-range correlation in inter-event durations becomes stronger over a longer time scale, i.e., when moving from a range of hours to days and further to months. We also observe interesting associations between the scaling exponent and a number of economic variables, in particular, in the inter-trade time series.
    Date: 2017–11

This nep-mst issue is ©2017 by Thanos Verousis. It is provided as is without any express or implied warranty. It may be freely redistributed in whole or in part for any purpose. If distributed in part, please include this notice.
General information on the NEP project can be found at For comments please write to the director of NEP, Marco Novarese at <>. Put “NEP” in the subject, otherwise your mail may be rejected.
NEP’s infrastructure is sponsored by the School of Economics and Finance of Massey University in New Zealand.