- Applied econometrics
- Damages estimation
- Energy market analysis
- Market manipulation
- Predictive modeling
- Time series econometrics
- Computer hardware and software
- Natural gas
- Chinese (Mandarin)
Ai Deng specializes in applied and theoretical econometrics. He has extensive experience in applying econometrics in a variety of litigation contexts—in particular, the estimation of damages and the establishment of liability in price-fixing cases. He also has experience in using statistical techniques, including machine learning/pattern recognition, to study alleged manipulation in financial markets.
Dr. Deng assists expert witnesses in all aspects of their empirical analyses, from performing and directing data processing to implementing state-of-the-art econometric techniques. He also assists attorney clients with understanding econometric arguments and issues put forward by both plaintiff and defendant experts, and with developing easy-to-understand documents that explain technical arguments to judges and juries. Within Bates White, Dr. Deng is responsible for overseeing the development of one of the firm’s proprietary analytical tools. His recent research, inspired by his consulting work, focuses on the econometrics of forecasting, model and variable selection, and structural changes.
Dr. Deng has published in multiple internationally renowned academic journals including Journal of Econometrics, Econometric Theory, Econometrics Journal, and Journal of Financial Econometrics. In addition, he is an active referee for a number of academic journals, including Journal of Econometrics, Journal of Business & Economic Statistics, Journal of Applied Econometrics, Journal of Empirical Finance, Journal of Time Series Econometrics, and Journal of Economic Surveys. Recently, Dr. Deng was invited to be an external expert to evaluate research programs to be funded by the Romanian National Council for Scientific Research.
- Conducted extensive analyses in support of cartel damages estimation for the expert witness in a number of litigations, including Oracle America, Inc. v. Micron Technology, Inc., In re TFT-LCD (Flat Panel) Antitrust Litigation, In re Urethane Antitrust Litigation, and In re Rubber Chemicals Antitrust Litigation.
- Assisted the expert witness and directed both the affirmative and rebuttal econometric analyses to examine NYMEX natural gas futures price behavior in In re Amaranth Advisors LLC.
- In In re Puerto Rico Jesús Trilla v. Commonwealth of Puerto Rico, a class-action lawsuit brought against fuel wholesalers, retailers, and the Department of Consumer Affairs in Puerto Rico, assisted the expert witness and developed multiple econometric tests to show that temperature adjustments in gasoline were fully passed through to final consumers.
- Conducted extensive econometric analyses proving defendant liability and calculating cartel damages in In re Dynamic Random Access Memory (DRAM) Antitrust Litigation. Analyses supported expert witnesses in both the affirmative and rebuttal reports.
- Developed rebuttal and affirmative econometric analyses to support multiple experts to analyze damages issues related to the misreporting of prices to publishers of natural gas price indices in In re Natural Gas Commodities Litigation.
PhD, Economics, Boston University
MA, Political Economy, Boston University
BA, Economics, Chongqing University
- Journal of Financial Econometrics, February 2014
- An Econometric Analysis of Recent Oil Price Movements: The Role of Political Events, Economic News, and Market FundamentalsJanuary 2012
- Local power of consistent tests for serial correlation against the nearly integrated, nearly white noise processEconomics Letters, April 2010
- Journal of the American Statistical Association, September 1, 2009
- Econometric Theory, June 2008
- A non-local perspective on the power properties of the Cusum and Cusum of Squares Tests for Structural ChangeJournal of Econometrics, January 2008
- A comparison of alternative asymptotic frameworks to analyse a structural change in a linear time trendThe Econometrics Journal, October 31, 2006
- Decomposing Predictability: A Multi-Resolution AnalysisLecture Series on Computer and Computational Sciences, April 2005