Quantifying preferential trading in the e-MID interbank market

Interbank markets allow credit institutions to exchange capital for purposes of liquidity management. These markets are among the most liquid markets in the financial system. However, liquidity of interbank markets dropped during the 2007–2008 financial crisis, and such a lack of liquidity influenced the entire economic system. In this paper, we analyse transaction data from the e-MID market which is the only electronic interbank market in the Euro Area and US, over a period of 11 years (1999–2009). We adapt a method developed to detect statistically validated links in a network, in order to reveal preferential trading in a directed network. Preferential trading between banks is detected by comparing empirically observed trading relationships with a null hypothesis that assumes random trading among banks doing a heterogeneous number of transactions. Preferential trading patterns are revealed at time windows of 3-maintenance periods. We show that preferential trading is observed throughout the whole period of analysis and that the number of preferential trading links does not show any significant trend in time, in spite of a decreasing trend in the number of pairs of banks making transactions. We observe that preferential trading connections typically involve large trading volumes. During the crisis, we also observe that transactions occurring between banks with a preferential connection occur at larger interest rates than the complement set—an effect that is not observed before the crisis.     

Modeling stock pinning

This paper investigates the effect of hedging strategies on the so-called pinning effect, i.e. the tendency of stock's prices to close near the strike price of heavily traded options as the expiration date nears. In the paper we extend the analysis of Avellaneda and Lipkin, who propose an explanation of stock pinning in terms of delta hedging strategies for long option positions. We adopt a model introduced by Frey and Stremme and show that, under the original assumptions of the model, pinning is driven by two effects: a hedging-dependent drift term that pushes the stock price toward the strike price and a hedging-dependent volatility term that constrains the stock price near the strike as it approaches it. Finally, we show that pinning can be generated by simulating trading in a double auction market. Pinning in the microstructure model is consistent with the Frey and Stremme model when both discrete hedging and stochastic impact are taken into account.     

Market microstructure,banks’ behaviour and interbank spreads: evidence after the crisis

We present a study of the European electronic interbank market of overnight lending (e-MID) before and after the beginning of the financial crisis. The main goal of the paper is to explain the structural changes of lending/borrowing features due to the liquidity turmoil. Unlike previous contributions that focused on banks’ dependent and macro information as explanatory variables, we address the role of banks’ behaviour and market microstructure as determinants of the credit spreads. We show that all banks experienced significant variations in their liquidity costs due to the sensitivity of interbank rates to the timing and side of trades. We argue that, while larger banks did experience better funding conditions after the crisis, this was not just a consequence of the “too big to fail” perception of the market. Larger banks have been able to play more strategically when managing their liquidity by taking advantage of the changing market microstructure.

     

Cross-correlation Measures in the High-frequency Domain

Abstract

On a high-frequency scale the time series are not homogeneous, therefore standard correlation measures cannot be directly applied to the raw data. To deal with this problem the time series have to be either homogenized through interpolation, or methods that can handle raw non-synchronous time series need to be employed. This paper compares two traditional methods that use interpolation with an alternative method applied directly to the actual time series. The three methods are tested on simulated data and actual trades time series.     

Introduction to the special issue on the 24th annual Workshop on Economic science with Heterogeneous Interacting Agents,London, 2019 (WEHIA 2019)

Journal of Economic Interaction and Coordination -