Intelligent Methods For Processing and Analysing Historical Order Book Data - The Case of Money Market Futures

26 January 2016
January 26, 2016

Time: 11:00 am
Location: Polo Scientifico e Tecnologico “Fabio Ferrari”, Room Garda, via Sommarive 5 - Povo (Trento)

Speaker 

  • Jing Nie, Durham University Business School

Abstract
 

Using the Eurodollar futures market as a case study, we have constructed a unique dataset consisting of both the inside quotes (best-bid, best-ask) and the entire activity within the limit order book. This allows us to classify, ex-post, the realized fraction of high frequency trading (HFT) activity within the market.

We then use this fraction as a dependent variable in a novel non-parametric regression to estimate the marginal effect on a series of market quality indicators. We demonstrate empirically that thresholds exist in terms of the impact of HFTs on market quality and this may explain the contradictory evidence found in prior studies of this type.

The objective of this talk is to introduce the background of trading mechanisms and some techniques on data extraction and manipulation, and provide some ideas for research on the algorithms of trades classification and detecting high frequency algorithm trading.

About the Speaker 

Jing Nie is a final year PhD candidate in finance at Durham University Business School in the United Kingdom. She holds a masters degree in Finance from the University of Leicester with a specialty in asset pricing and portfolio management. Jing’s primary area of research is empirical market microstructure, with a specific application on interest rate derivatives. Jings work on transaction level data analytics uses parallel computing and in-memory databases to analyze the informativeness of quoting and trades on the market. Jing's current research interests include: 1) market qualities and the factors influencing market qualities;  2) high-frequency price efficiency on money market futures; 3) term structure of trading on Eurodollar future markets. 

Contact Person Regarding this Talk: Fabio Massacci - fabio.massacci [at] unitn.it