TRADING AND DATA SCIENCE HAO NI OXFORD-MAN INSTITUTE OF - - PowerPoint PPT Presentation

ALGORITHMIC TRADING AND DATA SCIENCE

HAO NI OXFORD-MAN INSTITUTE OF QUANTITATIVE FINANCE

STEREOTYPES OF BANKERS AND SCIENTISTS

THE EVOLUTION OF TRADING VENUE

Algorithmic Trading

It encompasses trading systems that are heavily reliant on complex mathematical formulas and high-speed, computer programs to determine trading strategies.

Data

• Source: Massive financial data streams
• Data collection

Model

• Quantify the real world problem
• Propose a robust and effective model to describe the underlying data streams

Method

• Explore hidden patterns behind massive data streams
• Make better prediction for the future market

Execution

• Place trades automatically

WHY ALGORITHMS HELPS TRADING?

High speed

• The ability to handle more volume of trades
• High speed execution

Advanced Learning techniques

• Explore hidden patterns behind massive data streams
• Make better prediction for the future market

Decrease human intervention

• Free of human emotions
• Eliminate manual errors, missed opportunities etc

EXAMPLE: PAIRS TRADING

Source: http://www.nasdaq.com/article/dont-be-fooled-by-the-fancy-name- statistical-arbitrage-is-a-simple-way-to-profit-cm254669

Source: http://htxpro.squarespace.com/blog/2014/10/26/the- math-of-pairs-trading-execution-part-i

2010 FLASH CRASH

Source: TABB group

MY RESEARCH: CHANGE-POINT PROBLEM

Input –Output Pair (X, Y) : Y ~ f(X) + e

Bayesian framework

• f is random
• Prior distribution: GP(m, K)
• Posterior distribution P( f | (Xi, Yi)): updated

based on the observations (Xi, Yi).

Change-point

• K is a region-switching type([3] ).

Application: Detect and Predict the structural change

in the correlation of financial time series.

“MODELERS’ HIPPOCRATIC OATH”

I will remember that I didn’t make the world and it does not satisfy my equations. I will never sacrifice reality for elegance without explaining why I have done so No will I give the people who use my model false comfort about its accuracy. Instead I will make explicit its assumptions and oversights. I understand that my work may have enormous effects on society and the economy, many of them are beyond my comprehension.

BIBLIOGRAPHY

[1] Scott Patterson, The Quants: How a New Breed of Math Whizzes Conquered Wall Street and Nearly Destroyed It; Crown Business, 2011. [2] Scott Patterson, Dark Pools: The rise of A.I. trading machines and the looming threat to Wall Street; Crown Business, 2013. [3] Garnett, Roman, et al. "Sequential Bayesian prediction in the presence of changepoints and faults." The Computer Journal (2010): bxq003.