Scribd
Upload
46 views34 pages

Algorithmic Trading Yves Algorithmic Trading Machine Deep Learning Data

Uploaded by

WLS
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
46 views34 pages

Algorithmic Trading Yves Algorithmic Trading Machine Deep Learning Data

Uploaded by

WLS
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 34

Algorithmic Trading

—With Machine & Deep Learning

ODSC Europe in London | 12. October 2017

Dr. Yves J. Hilpisch


“Pichai said that as an ‘AI first’ company, this is a ‘unique
moment in time’ for Google to combine hardware, software
and artificial intelligence. ‘It's radically rethinking how
computing should work’, he said."

Business Standard, "Google Ramps up Hardware Business", 06. October 2017.


Introduction
http://tpq.io
http://pqp.io
http://hilpisch.com
http://books.tpq.io
http://certificate.tpq.io
Link to Study Plan
http://pyalgo.tpq.io
http://hilpisch.com/bcioo_tpq.pdf
http://hilpisch.com/tpq_silicon_review.pdf
Machine Learning & Algorithmic Trading
machine & deep learning

optimization, prediction
data training & (“self-driving car”)
learning
algorithms automation
testing
hardware trading
validation (“money making
machine”)

algorithmic trading
Emulation

powerful complete market


hardware & replication with
software all agents
Markets &
Algorithms
Agents

x x

f(x)

y y
Agenda
TOPICS FOR TODAY

1. Reading Financial Time Series Data with pandas


2. Formulating a Financial Trading Strategy
3. Vectorized Backtesting of the Trading Strategy
4. Random Walk Hypothesis
5. Prediction based on Classification Algorithm
6. Prediction based on Neural Nets
7. Prediction based on Deep Neural Network
Interactive style as a major element
“Making mistakes together.”
Skill-based workshop approach
From Anders Ericsson (2016): Peak — Secrets from the New Science of Expertise.
The Booley Head, London.

“When you look at how people are trained in the professional and business
worlds, you find a tendency to focus on knowledge at the expense of skills.”

“... I believe the best approach will be to develop skills-based training programs
that will supplement or completely replace the knowledge-based approaches
that are the norm now in many places. The strategy acknowledges that because
what is ultimately most important is what people are to do,
training should focus on doing rather than on knowing.”
The Gist
http://gist.github.com/yhilpisch

https://goo.gl/LAQ8Ze
The Data
We work with EOD data for the EUR/USD rate
Data Source: FXCM Financial Capital Markets Ltd.

http://hilpisch.com/eurusd.csv
The Benchmark Case
of Random Walks
“For many years, economists, statisticians, and teachers of finance have been
interested in developing and testing models of stock price behavior. One important
model that has evolved from this research is the theory of random walks. This theory
casts serious doubt on many other methods for describing and predicting stock price
behavior—methods that have considerable popularity outside the academic world.
For example, we shall see later that, if the random-walk theory is an accurate
description of reality, then the various “technical” or “chartist” procedures for
predicting stock prices are completely without value.”

Eugene F. Fama (1965): “Random Walks in Stock Market Prices”.


“A market is efficient with respect to an information set S if it is impossible to make
economic profits by trading on the basis of information set S.”

Michael Jensen (1978): “Some Anomalous Evidence Regarding Market Efficiency”.


If a the price of a financial instrument follows a
(simple) random walk
(no drift & normally distributed returns),
then it rises and falls with
the same probability of 50% (“toss of a coin”).

In such a case,
the best predictor of tomorrow’s stock price
—in a least-squares sense—
is today’s stock price.
Outlook
singularity

exponential forces at work:


• technology improvements
• capital accumulation
• talent accumulation
• network effects
The Python Quants GmbH
Dr. Yves J. Hilpisch
+49 3212 112 9194
http://tpq.io | [email protected]
@dyjh

You might also like