|
on Market Microstructure |
| By: | Hans Degryse; Mark Van Achter; Gunther Wuyts |
| Abstract: | We present a dynamic microstructure model where a dealer market (DM) and a crossing network (CN) interact. We consider sequentially arriving agents having different valuations for an asset. Agents maximize their profits by either trading at a DM or by submitting an order for (possibly) uncertain execution at a CN. We develop the analysis for three different informational settings: transparency, “complete” opaqueness of all order flow, and “partial” opaqueness (with observable DM trades). We find that a CN and a DM cater for different types of traders. Investors with a high eagerness to trade are more likely to prefer a DM. The introduction of a CN increases overall order flow by attracting traders who would not otherwise submit orders (“order creation”). It also diverts trades from the DM. The transparency and “partial” opaqueness settings generate systematic patterns in order flow. With transparency, the probability of observing a CN order at the same side of the market is smaller after such an order than if it was not. Buy (sell) orders at a CN are also less likely to attract subsequent sell (buy) orders at the DM. |
| Date: | 2008–03 |
| URL: | http://d.repec.org/n?u=RePEc:ete:ceswps:ces0415&r=mst |
| By: | Hans Degryse; Mark Van Achter |
| Abstract: | A diversity of so-called Alternative Trading Systems (ATS) has challenged the existing traditional exchanges. This paper studies the impact of these ATS on the liquidity on the traditional financial markets using a market microstructure approach. In the United States ATS have been particularly successful in attracting trade in the Nasdaq dealer market, whereas they are less successful in competition with the NYSE. The theoretical reasoning behind this conjecture is that the agency nature of trading at the ATS allows investors to trade directly with each other without the intervention of a dealer. We argue that since continental European exchanges are typically organized as auction markets implying an agency nature of trading, the liquidity externality will prevent the auction-type ATS from breaking through and acquiring a significant market share in Europe. Only Crossing Networks may turn out to be more successful in realizing trades in Europe as they rely on the efficiency of price discovery on their primary market. |
| Date: | 2008–03 |
| URL: | http://d.repec.org/n?u=RePEc:ete:ceswps:ces0122&r=mst |
| By: | Fischer, Andreas M; Ranaldo, Angelo |
| Abstract: | Does global currency volume increase on days when the Federal Open Market Committee (FOMC) meets? To test the hypothesis of excess currency volume on FOMC days, we use a novel data set from the Continuous Linked Settlement (CLS) Bank. The CLS measure captures roughly half of the global trading volume in foreign exchange (FX) markets. We find strong evidence that trading volume increases in the order of 5% across currency areas on FOMC days during 2003 to 2007. This result holds irrespective of the size of price changes in currency markets and FOMC policy shocks. The new evidence of excess FX trading on FOMC days is inconsistent with standard models of the asset market approach with homogenous agents. |
| Keywords: | FOMC; Global linkages; Trading volume |
| JEL: | F31 G12 |
| Date: | 2008–03 |
| URL: | http://d.repec.org/n?u=RePEc:cpr:ceprdp:6753&r=mst |
| By: | Martínez de Albeniz, Victor (IESE Business School); Vendrell, Josep M. (IESE Business School) |
| Abstract: | In this paper we consider the problem of a trader who purchases a commodity in one market and resells it in another. The trader is capacitated: the trading volume is limited by operational constraints, e.g., logistics. The two markets quote different prices, but the spread is reduced when trading takes place. We are interested in finding the optimal trading policy across the markets so as to obtain the maximum profit in the long-term, taking into account that the trading activity influences the price processes, i.e., market power. As in the no-market-power case, we find that the optimal policy is determined by three regions, where 1) move as much as possible from one market to the other; 2) the same in the opposite direction; or 3) do nothing. Finally, we use the model to analyze kerosene price differences between New York and Los Angeles. |
| Keywords: | commodity trading; price processes; inventory management; |
| Date: | 2008–01–07 |
| URL: | http://d.repec.org/n?u=RePEc:ebg:iesewp:d-0728&r=mst |
| By: | Jongen, Ron; Verschoor, Willem F C; Wolff, Christian C; Zwinkels, Remco C.J. |
| Abstract: | This paper analyzes the sources of the differential beliefs of market participants in the foreign exchange market and their relative role in forming exchange rate expectations. We find that there are distinct periods of high and low dispersion and document that dispersion arises because of a combined effect of market participants holding individual information and attach different weights to some elements of the common information set. In addition to these two effect, we also document evidence of the existence of different types of agents and find that chartist rules are predominantly used at the shorter spectrum of the forecast horizon and fundamentalist rules are predominantly used at the longer spectrum of the forecast horizon. Finally, our evidence suggests that the relationship between market volatility and trader dispersion tends to be significant and positive for different measures of both trader heterogeneity and market volatility. |
| Keywords: | exchange rates; expectations; heterogeneity; survey data |
| JEL: | F31 |
| Date: | 2008–03 |
| URL: | http://d.repec.org/n?u=RePEc:cpr:ceprdp:6738&r=mst |
| By: | Pierre Bajgrowicz (University of Geneva); Olivier Scaillet (University of Geneva and Swiss Finance Institute) |
| Abstract: | We revisit the apparent historical success of technical trading rules on daily prices of the Dow Jones index. First, we use the False Discovery Rate as a new approach to data snooping. The advantage of the FDR over existing methods is that it is more powerful and not restricted only to the best rule in the sample. Second, we perform persistence tests and conclude that an investor would not have been able to select ex ante the future best-performing rules. Finally, we show that the performance fully disappears once transaction costs are taken into account. |
| Keywords: | Technical Trading, False Discovery Rate,Persistence, Transaction Costs. |
| JEL: | C12 C15 G11 G14 |
| Date: | 2007–05 |
| URL: | http://d.repec.org/n?u=RePEc:chf:rpseri:rp0805&r=mst |
| By: | Olivier Vergote |
| Abstract: | This paper proposes a new procedure to determine the time of the prevailing quote relative to the time of the trade for NYSE stock data obtained from the TAQ database. The procedure tests whether the quote revision frequency around a trade is contaminated by quote revisions triggered by a trade, and then determines the smallest timing adjustment needed to eliminate this contamination. An application to various stocks and sample periods shows that the time difference between trade and quote reporting lags varies across stocks and time. The procedure takes this variation into account and hence offers a stock- and time-specific update to the Lee and Ready (1991) 5-second rule. |
| Date: | 2008–03 |
| URL: | http://d.repec.org/n?u=RePEc:ete:ceswps:ces0510&r=mst |