Adverse-Selection Costs and the Probability of Information-Based Trading

Prior studies offer various empirical models to decompose the observed bid‐ask spread into the adverse‐selection and transitory (order‐processing and inventory‐holding) components. There is limited evidence, however, on whether the spread components estimated from these models indeed measure what they purport to measure. In this study, we show that the estimates of the adverse‐selection component given by these models are positively and significantly related to the probability of information‐based trading (PIN), after controlling for the endogeneity of the PIN and other stock attributes. These results provide direct empirical support for the spread component models examined in the present study.     

Earnings Announcements and the Components of the Bid-Ask Spread

This study investigates the behavior of the components of the bid-ask spread around earnings announcements. We find that the adverse selection cost component significantly increases surrounding the announcements, while the inventory holding and order processing components significantly decline during the same periods. Our results suggest that the directional change in the total bid-ask spread depends on the relative magnitudes of the changes in these three components. Specifically, the decreases in inventory holding costs and order processing costs imply that earnings announcements may have an insignificant impact on the total bid-ask spread, even when they result in increased information asymmetry.     

Inferring the Components of the Bid-Ask Spread: Theory and Empirical Tests

The relation between the square of the quoted bid-ask spread and two serial covariances—the serial covariance of transaction returns and the serial covariance of quoted returns—is modeled as a function of the probability of a price reversal, π, and the magnitude of a price change, ?, where ? is stated as a fraction of the quoted spread. Different models of the spread are contrasted in terms of the parameters, π and ?. Using data on the transaction prices and price quotations for NASDAQ/NMS stocks, π and ? are estimated and the relative importance of the components of the quoted spread—adverse information costs, order processing costs, and inventory holding costs—is determined.     

沪深两市信息非对称程度及其对比:基于MRR方法的经验研究

本文采用Madhavan,Richarson and Roomans(1997)模型对沪深两市A股市场的信息非对称程度进行了经验比较.首先将隐性价差分解为逆向选择成本和指令处理成本两部分,然后以隐性价差中逆向选择成本所占的比重作为市场信息非对称程度的衡量指标,最后以该衡量指标实证比较了沪深两市的信息非对称的日内变动模式和大小.经验分析表明,沪市信息非对称程度的日内变动模式呈倒"U"形,而深市信息非对称程度的日内模式无明显规律;与沪市类似,深市信息非对称在开盘初期呈上升趋势和午市连续竞价的末期呈下降趋势;沪市的信息非对称程度在大多数交易时段要大于深市.     

Government intervention and adverse selection costs in foreign exchange markets

An important group of traders in the foreign exchange marketis governments who often adhere to a foreign exchange rate policyof occasional interventions with otherwise floating rates. Inthis article we provide a theoretical model and empirical evidencethat government foreign exchange interventions create significantadverse selection problems for dealers. In particular, our modelshows that the adverse selection component of the foreign exchangespread is positively related to the variance of unexpected interventionand that expected intervention has no impact on the spread.After controlling for inventory and order processing costs,we find that bid-ask spreads increase with U.S. dollar and Germandeutsche mark foreign exchange rate intervention during theperiod 1976-1994. Furthermore, when the intervention is decomposedinto expected and unexpected components, we find a statisticallyand economically significant increase in spreads with the varianceof unexpected intervention, while expected intervention hasno significant impact on spreads.     

Modeling the bid/ask spread: measuring the inventory-holding premium

The need to understand and measure the determinants of market maker bid/ask spreads is crucial in evaluating the merits of competing market structures and the fairness of market maker rents. This study develops a simple, parsimonious model for the market maker's spread that accounts for the effects of price discreteness induced by minimum tick size, order-processing costs, inventory-holding costs, adverse selection, and competition. The inventory-holding and adverse selection cost components of spread are modeled as an option with a stochastic time to expiration. This inventory-holding premium embedded in the spread represents compensation for the price risk borne by the market maker while the security is held in inventory. The premium is partitioned in such a way that the inventory-holding and adverse selection cost components, as well as the probability of an informed trade, are identified. The model is tested empirically using Nasdaq stocks in three distinct minimum tick size regimes and is shown to perform well both in an absolute sense and relative to competing specifications.     

Bid-ask spread dynamics in foreign exchange markets

The aim of this paper is to examine the short term dynamics of foreign exchange rate spreads. Using a vector autoregressive model (VAR) we show that most of the variation in the spread comes from the long run dependencies between past and future spreads rather than being caused by changes in inventory, adverse selection, cost of carry or order processing costs. We apply the Integrated Cumulative Sum of Squares (ICSS) algorithm of Inclan and Tiao (1994) to discover how often spread volatility changes. We find that spread volatility shifts are relatively uncommon and shifts in one currency spread tend not to spillover to other currency spreads.     

Competitive inventory management in Treasury markets

We decompose US Treasury bid-ask spreads into inventory, adverse selection and order processing costs by using the fact that inventory trades have different effects on spreads than do proprietary trades. We exploit this asymmetry and develop a technique to identify the three components of the spread in order to test three hypotheses: dealers make larger changes to inventory (1) following macroeconomic announcements (2) at the start and toward the end of the New York trading hours, and (3) when transaction sizes are relatively large. We test these predictions using GovPX data for on-the-run 2-year and 10-year Treasury Notes. All three predictions are supported. We also assess how primary dealers react to the Federal Reserve’s open market operations (OMOs). Our findings reveal interesting intraday patterns in the inventory component for both securities.     

The components of the bid-ask spread: a general approach

A simple time-series market microstructure model is constructedwithin which existing models of spread components are reconciled.We show that existing models fail to decompose the spread intoall its components. Two alternative extensions of the simplemodel are developed to identify all the components of the spreadand to estimate the spread at which trades occur. The empiricalresults support the presence of a large order processing componentand smaller, albeit significant, adverse selection and inventorycomponents. The spread components differ significantly accordingto trade size and are also sensitive to assumptions about therelation between orders and trades.     

Specialists,Limit‐Order Traders,and the Components of the Bid‐Ask Spread

This study compares the components of the bid‐ask spread estimated from quotes that reflect the trading interest of specialists with those estimated from limit‐order quotes and all available quotes for a sample of New York Stock Exchange (NYSE) stocks. The results show that the adverse selection component of the spread estimated from specialist quotes is significantly smaller than the corresponding figures from limit‐order quotes and entire quotes. We interpret this as evidence that NYSE specialists transfer at least a part of adverse selection costs to outsiders through the discretionary use of limit orders. Our results show that the estimation/interpretation of the components of the spread using quote data that include both specialist and limit‐order interests is problematic.     

Determinants of the components of bid-ask spreads on stocks

In this paper we show that George et al. (GKN, 1991) estimators of the adverse selection and order processing cost components of the bid-ask spread are biased due to intertemporal variations in the bid-ask spread. We use alternative estimators that correct this bias and that are applicable to individual securities, and estimate these cost components empirically using data on NYSE/AMEX stocks. As expected, our results indicate that on average adverse selection costs account for approximately 50% of the bid-ask spread, sharply higher than the estimates of 8-10% obtained by GKN for NASDAQ stocks and 21% that we obtain for NYSE/AMEX stocks using GKN's estimators. We then conduct cross-sectional regressions designed primarily to determine whether adverse selection costs vary across specialists after controlling for firm size and other factors. Consistent with previously established hypotheses, we find that adverse-selection costs vary across specialists, and that this variation is related to the number of securities that the specialist handles.     

ADVERSE SELECTION,INVENTORY‐HOLDING COSTS,AND DEPTH

We examine whether specialist depth quotes are related to the adverse‐selection and inventory‐holding‐cost components of the spread. Consistent with theory that predicts an inverse relation between depths and informed trading risk, we find that depth quotes are strongly inversely related to the adverse‐selection component of the spread. We also find that depth quotes are inversely related to the inventory‐holding‐cost component, though this relation is weaker than the relation between depths and adverse selection. Our evidence suggests cross‐sectional variation in depths is driven primarily by variation in informed trading risk, as proxied for by the adverse‐selection component, rather than by inventory concerns. JEL classification: G10, G14     

A pairs trading strategy based on linear state space models and the Kalman filter

Among many strategies for financial trading, pairs trading has played an important role in practical and academic frameworks. Loosely speaking, it involves a statistical arbitrage tool for identifying and exploiting the inefficiencies of two long-term, related financial assets. When a significant deviation from this equilibrium is observed, a profit might result. In this paper, we propose a pairs trading strategy entirely based on linear state space models designed for modelling the spread formed with a pair of assets. Once an adequate state space model for the spread is estimated, we use the Kalman filter to calculate conditional probabilities that the spread will return to its long-term mean. The strategy is activated upon large values of these conditional probabilities: the spread is bought or sold accordingly. Two applications with real data from the US and Brazilian markets are offered, and even though they probably rely on limited evidence, they already indicate that a very basic portfolio consisting of a sole spread outperforms some of the main market benchmarks.     

Instability of return prediction models

This study examines evidence of instability in models of ex post predictable components in stock returns related to structural breaks in the coefficients of state variables such as the lagged dividend yield, short interest rate, term spread and default premium. We estimate linear models of excess returns for a set of international equity indices and test for stability of the estimated regression parameters. There is evidence of instability for the vast majority of countries. Breaks do not generally appear to be uniform in time: different countries experience breaks at different times. For the majority of international indices, the predictable component in stock returns appears to have diminished following the most recent break. We assess the adequacy of the break tests and model selection procedures in a set of Monte Carlo experiments.     

The determinants of market-wide issue cycles for initial public offerings

This paper identifies the determinants of market-wide issue cycles for initial public offerings (IPOs) using an autoregressive conditional count model. We consider whether IPO volume is related to business conditions, investor sentiment, and time variation in adverse selection costs caused by asymmetric information between managers and investors. We provide evidence indicating that time variation in business conditions and investor sentiment are important determinants of monthly issue activity. By contrast, time variation in adverse selection costs does not significantly affect IPO volume.     

A SPECIALIST'S QUOTED DEPTH AS A STRATEGIC CHOICE VARIABLE: AN APPLICATION TO SPREAD DECOMPOSITION MODELS

Although there is a sizable literature demonstrating that liquidity and transaction costs are multidimensional, researchers continue to estimate adverse‐selection costs using only prices. We present a model of a profit‐maximizing specialist who posts prices and depths. The model is simulated to measure changes in the adverse‐selection component of the spread that result under different levels of informed trading. We find that spread decompositions fail to capture the full extent of adverse‐selection risk when specialists choose depth. We recommend that researchers use adverse‐selection measures that account for depth as well as spread to mitigate this problem.     

The Effect of Decimalization on Trade Size and Adverse Selection Costs

Abstract:   We examine adverse selection costs around NYSE decimalization. Further, we analyze the relation between adverse selection costs and trade size. We find a significant increase in the percentage adverse selection cost and a reduction in dollar adverse selection cost (percentage adverse selection multiplied by the spread) following complete decimalization on the NYSE. On estimating the adverse selection components by trade size classes, we find a decline in dollar adverse selection costs in trades of all sizes, with the strongest evidence coming from medium size trades, followed by small and large size trades. One implication of our findings is that there appears to be less stealth trading following complete decimalization and less institutional trading overall.     

PERFORMANCE OF STOLL's SPREAD COMPONENT ESTIMATOR: EVIDENCE FROM SIMULATIONS,TIME-SERIES,AND CROSS-SECTIONAL DATA

Stoll (1989) introduces an intuitive procedure to estimate the basic components of the bid-ask spread (order-processing cost, inventory cost, and adverse-selection cost). He also provides reasonable estimates of the magnitudes of the order-processing, inventory, and adverse-selection costs of making markets for a large cross-section of NASDAQ/NMS stocks. Empirical applications of Stoll's model produce widely different estimates of the bid-ask spread components. We derive the sampling properties of Stoll's estimator of the realized bid-ask spread, i.e., the sum of the order-processing and inventory components. We test Stoll's model in simulations, using the ideal conditions implied by the model. We conclude that noise in serial covariance estimates causes estimates of the realized spread to be severely biased and highly unreliable in short time-series and small cross-sectional samples.     

Selection Bias and Auditing Policies for Insurance Claims

Selection bias results from a discrepancy between the range of estimation of a statistical model and its range of application. This is the case for fraud risk models, which are estimated on audited claims but applied on incoming claims in the design of auditing strategies. Now audited claims are a minority within the parent sample since they are chosen after a severe selection performed by claims adjusters. This article presents a statistical approach that counteracts selection bias without using a random auditing strategy. A two‐equation model on audit and fraud (a bivariate probit model with censoring) is estimated on a sample of claims where the experts are left free to take the audit decision. The expected overestimation of fraud risk derived from a single‐equation model is corrected. Results are close to those obtained with a random auditing strategy, at the expense of some instability with respect to the regression components set. Then we compare auditing policies derived from the different approaches.     

二元Copula函数的投资组合模型选择

Coplua模型是组合投资风险评估中常用模型,它具有多种不同的类型,模型选择的好坏对风险评估结果具有至关重要的影响.本文主要比较了二元正态Copula模型和二元t-Copula模型对中国股市数据拟合的优劣程度.针对这两种模型,利用上证综指、深证成指、上证基金、深证基金、东风汽车、中国石化、宝钢股份和万家乐的日收盘价数据估计相应的参数得到相应的拟合分布,然后分别与经验Copula函数作比较,通过计算拟合分布与经验分布之间的距离,得出二元t-Cop-ula函数能更好地拟合两组投资组合的日收益率数据的结论.