a vision for quantitative investing in the data economy by michael beal at quantcon 2016

30
THE FUTURE OF INVESTING IN THE DATA ECONOMY 0

Upload: quantopian

Post on 16-Apr-2017

316 views

Category:

Economy & Finance


0 download

TRANSCRIPT

T H E F U T U R E O F I N V E S T I N G I N T H E D ATA E C O N O M Y

0

TAB L E O F C O N TE N TS

Extremely Confidential. Not for Public Distribution. | DATA CAPITAL MANAGEMENT | 1

QuantCon 2016

Saturday, April 9, 2016

Michael M. Beal

Got API?

[email protected]

THE FUTURE OF INVESTING IN THE DATA ECONOMY

L E S S O N S F R O M TH E I N D U S TR I AL R E V O L U TI O N

Extremely Confidential. Not for Public Distribution. | DATA CAPITAL MANAGEMENT | 2

Suggestions for Start-ups in the Data Economy

• Standardize industry taxonomies

• Be the best at depth of information for a given vertical

• Take responsibility for the veracity of information

• Focus on speed of delivery

• Focus on historical consistency

• Make adjustments visible to upstream users

• Focus on permissible data use rights

• Don’t be all things to all people

Got API?

[email protected]

Our goal is to help drive a standard for other players in our ecosystem to coalesce around

• This approach helps avoid the “tragedy of the commons” and maximize collective ROI

• We are set up to be “early customers” of new technologies / beta releases

• We drive a culture of focused and targeted feedback

For those with self-interests congruent to ours; let’s help each other

TAB L E O F C O N TE N TS

Extremely Confidential. Not for Public Distribution. | DATA CAPITAL MANAGEMENT | 3

DATA CAPITAL MANAGEMENT OVERVIEW

WHAT IS THE DATA ECONOMY?

WHAT NEW CHALLENGES DOES THE DATA ECONOMY CREATE FOR INVESTORS?

THE FUTURE OF QUANTITATIVE INVESTING IN THE DATA ECONOMY

LESSONS FROM THE INDUSTRIAL REVOLUTION

Data Capital Management (“DCM”), is a news-aware systematic hedge

fund. DCM specializes in strategies that make use of novel sources of

information (news, images, social networks data, weather, etc.) as well

as state of the art machine learning analytics to generate differentiated

and uncorrelated investment returns.

We are a team of PhDs and MBAs, with degrees in computer science,

engineering, mathematics and business management, with expertise in

quantitative research and risk management at global investment banks,

fundamental analysis at leading investment funds and cutting edge

technologies in big data companies.

DCM’s proprietary trading system delivers uncorrelated, market-neutral

returns in liquid public markets through a portfolio of fully-systematic

advanced strategies.

Extremely Confidential. Not for Public Distribution. | DATA CAPITAL MANAGEMENT | 4

I N TR O D U C T I O N TO D ATA C AP I TA L M AN AG E M E N T

Extremely Confidential. Not for Public Distribution. | DATA CAPITAL MANAGEMENT | 5

D C M N E W S - AWAR E I N V E S TM E N T P H I L O S O P H Y

Fundamental AnalysisThe analysis of company business

drivers and macroeconomic factors

-25

Analysis of

3,000+ listed

stocks & ETFs

in real time

Real Time Analysis at Large Scale andAutomatic Trade Execution

Quantitative AnalysisThe analysis of security price and

trading volume statistics

Novel Data Sources AnalysisCutting edge machine-learning

algorithms to analyze novel data

sources such as news, images, etc.

DCM combines the rules of fundamental investing, quantitative analysis, and trading

signals derived from novel data sources and machine learning algorithms to identify

alpha catalysts to changes in security prices

Parallel Cognitive Computing

P R I C E S R E AC T Q U I C K E R TH AN P E O P L E

Extremely Confidential. Not for Public Distribution. | DATA CAPITAL MANAGEMENT | 6

DCM’s proprietary technology enables a faster response to changing prices and unexpected news announcements.

Source: Interactive Brokers price data.

For illustrative purposes only. Back test results are not indicative of future returns. Strategies are preliminary and are not necessarily those that will be deployed in the market.

Source: Bloomberg news, Interactive Brokers price data.

For illustrative purposes only. Back test results are not indicative of future returns. Strategies are preliminary and are not necessarily those that will be deployed in the market.

10.3%

-12.3%-13.8%

-20.0%

-15.0%

-10.0%

-5.0%

0.0%

5.0%

10.0%

15.0%

Oil (WTI) Chipotle(CMG)

Staples(SPLS)

For Illustrative Purposes Only. Back test results are not indicative of future returns. Strategies are preliminary and are not necessarily those that will be deployed in the market.

• 8.27.15: Venezuela

seeks OPEC emergency

meeting on oil prices

• 11.20.15: Oregon

agency probes E. coli

cases linked to Chipotle

• 12.7.15: Staples-Office

Depot Deal in Doubt as

FTC Moves to Block

A

B

C

A B C

1-day price returns following unexpected news events

I N V E S TM E N T P H I L O S O P H Y

Extremely Confidential. Not for Public Distribution. | DATA CAPITAL MANAGEMENT | 7

DCM combines the rules of fundamental investing, quantitative analysis, and trading signals derived from novel data sources and

machine learning algorithms to identify short-term-alpha catalysts to changes in security prices

• Temporal

• Cross-sectional

• Short-term stat-arb

• Rebalancing

• Momentum

• Momentum Reversals

• Machine Learning Predictive

• Corporate Earnings

• Central Bank

• Cross-asset arbitrage

• M&A arbitrage

• News based

News-aware

investing enables

greater adaptive

diversification to

varying market

conditions through

an ability to switch

between investment

models based on

“market psychology”

Mean Reversion

Trend Following (Momentum)

Event Based

Str

ate

gy C

ate

go

rie

s

Traditional portfolio theory

Price Volume

TechnicalExecution

Flow

Consumer

Net Exports

Government News Satellite Sensors

SpeechesWeather

Earnings & Margins

Valuations & Growth

Data Categories

Price / Volume Macro Economics Fundamental Novel Data

Etc.

News-aware investing

Commodities

TAB L E O F C O N TE N TS

Extremely Confidential. Not for Public Distribution. | DATA CAPITAL MANAGEMENT | 8

DATA CAPITAL MANAGEMENT OVERVIEW

WHAT IS THE DATA ECONOMY?

WHAT NEW CHALLENGES DOES THE DATA ECONOMY CREATE FOR INVESTORS?

THE FUTURE OF QUANTITATIVE INVESTING IN THE DATA ECONOMY

LESSONS FROM THE INDUSTRIAL REVOLUTION

W H AT I S TH E D ATA E C O N O M Y?

Extremely Confidential. Not for Public Distribution. | DATA CAPITAL MANAGEMENT | 9

The Data Economy is the economy in

which individuals, institutions and

corporations commercialize their

Intellectual Property assets and

services in the form of web-based APIs

to third parties with the goal of

monetizing Positive Data Externalities

through the creation of new information-

based assets

Source: google images

For illustrative purposes only. Back test results are not indicative of future returns.

W H AT I S TH E D ATA E C O N O M Y?

Extremely Confidential. Not for Public Distribution. | DATA CAPITAL MANAGEMENT | 10

The Next Industrial

Revolution:

An Economy Based on

the Creation and

Exchange of Data

TAB L E O F C O N TE N TS

Extremely Confidential. Not for Public Distribution. | DATA CAPITAL MANAGEMENT | 11

DATA CAPITAL MANAGEMENT OVERVIEW

WHAT IS THE DATA ECONOMY?

WHAT NEW CHALLENGES DOES THE DATA ECONOMY CREATE FOR INVESTORS?

THE FUTURE OF QUANTITATIVE INVESTING IN THE DATA ECONOMY

LESSONS FROM THE INDUSTRIAL REVOLUTION

N E W D ATA E C O N O M Y C H AL L E N G E S F O R TR AD I T I O N A L I N V E S TO R S

Extremely Confidential. Not for Public Distribution. | DATA CAPITAL MANAGEMENT | 12

The new investment world requires an increased ability to handle breadth depth and

complication of relationships….

(1)Breadth: Identification of most relevant data feeds and information sources (news, images,

social networks data, macroeconomic data, etc.) and linking them to all the Entities

(companies, currencies, commodities) they impact. Real-time analysis of massive amounts

of heterogeneous data and predicting their impact on related Entities.

(2)Depth: 360° Entity Relationships Graph modeling relationships between vast number of

related entities (e.g. customer-supplier relationships) and their dependencies. Real-time

monitoring of changes on an entity level and predicting the impact on related entities.

(3)Speed: Identify, Analyze and execute long/short strategies based on new information (e.g.

change in relevant data, or change on an entity level) in under 5 seconds.

(4) Experience: Ability to understand macro and micro regimes and their likely impact on

company cash flows and investment returns

While data has increased

in value; the relative

importance of any given

data point on its own has

decreased

N E W D ATA E C O N O M Y C H AL L E N G E S F O R TR AD I T I O N A L I N V E S TO R S

Extremely Confidential. Not for Public Distribution. | DATA CAPITAL MANAGEMENT | 13

Breadth Depth Speed of Data has exploded

Source: DCM Analytics

For illustrative purposes only. Back test results are not indicative of future returns.

Extremely Confidential. Not for Public Distribution. | DATA CAPITAL MANAGEMENT | 14

N E W D ATA E C O N O M Y C H AL L E N G E S F O R TR AD I T I O N A L I N V E S TO R S

Company networks have gotten more complex; making defined SQL queries

increasingly painful…

Source: google images

For illustrative purposes only. Back test results are not indicative of future returns.

Extremely Confidential. Not for Public Distribution. | DATA CAPITAL MANAGEMENT | 15

N E W D ATA E C O N O M Y C H AL L E N G E S F O R TR AD I T I O N A L I N V E S TO R S

… A machine learning approach is required to understand complex entity relationships

Source: google images

For illustrative purposes only. Back test results are not indicative of future returns.

N E W D ATA E C O N O M Y C H AL L E N G E S F O R TR AD I T I O N A L I N V E S TO R S

Extremely Confidential. Not for Public Distribution. | DATA CAPITAL MANAGEMENT | 16

… Yet as humans gain experience, processing power decreases

Source: google images

For illustrative purposes only. Back test results are not indicative of future returns.

N E W D ATA E C O N O M Y C H AL L E N G E S F O R TR AD I T I O N A L I N V E S TO R S

Extremely Confidential. Not for Public Distribution. | DATA CAPITAL MANAGEMENT | 17

In a world driven by trend seeking algorithms, are fundamental long/short investors prepared for

the data economy?

We have come regretfully to the

conclusion that the current

algorithmically driven market

environment is one which is

increasingly incompatible with

our fundamental, research

orientated, investment process

– Nevsky Capital

Source: google images

For illustrative purposes only. Back test results are not indicative of future returns.

TAB L E O F C O N TE N TS

Extremely Confidential. Not for Public Distribution. | DATA CAPITAL MANAGEMENT | 18

DATA CAPITAL MANAGEMENT OVERVIEW

WHAT IS THE DATA ECONOMY?

WHAT NEW CHALLENGES DOES THE DATA ECONOMY CREATE FOR INVESTORS?

THE FUTURE OF QUANTITATIVE INVESTING IN THE DATA ECONOMY

LESSONS FROM THE INDUSTRIAL REVOLUTION

| DATA CAPITAL MANAGEMENT | 19

In essence all professional investment managers share the same process to make decisions

DATA ACQUISITION ANALYSIS DECISION

Breadth: Data is data; novel

data sources including news,

images, social networks,

macroeconomic feeds, etc. are

linked to price movements in

securities, currencies,

commodities, etc.

Depth: Analyzes, prioritizes

and monitors big-data input

and its impact using systematic

and quantitative metrics

Speed: Real-time, rules-based

extraction and interpretation of

information based on event

triggers from over 20,000

leading global newswires,

online newspapers,

aggregators,

and blogs in under 5 seconds

VOLUME &

VARIETYVERACITY VELOCITY

F U TU R E O F Q U AN T I TAT I V E I N V E S TI N G I N TH E D ATA E C O N O M Y

Extremely Confidential. Not for Public Distribution. | DATA CAPITAL MANAGEMENT | 20

• Parallel Processing

• In-memory compute

• Elastic Resource Management

• On-Demand load provisioning

Advances from the west coast have enabled machines to scale horizontally and linearly

Source: DCM Analytics; Google images

For illustrative purposes only. Back test results are not indicative of future returns.

F U TU R E O F Q U AN T I TAT I V E I N V E S TI N G I N TH E D ATA E C O N O M Y

Extremely Confidential. Not for Public Distribution. | DATA CAPITAL MANAGEMENT | 21

Cognitive Computing is in the early stages; the capital markets afford a fertile ground for

rapid learning

Source: IBM, DCM and google images

For illustrative purposes only. Back test results are not indicative of future returns.

F U TU R E O F Q U AN T I TAT I V E I N V E S TI N G I N TH E D ATA E C O N O M Y

-25

Analysis of

3,000+ listed

stocks & ETFs

in real time

Extremely Confidential. Not for Public Distribution. | DATA CAPITAL MANAGEMENT | 22

Machine learning is impacting all industries; in the process algorythms are learning

generalized and industry-specific knowledge

F U TU R E O F Q U AN T I TAT I V E I N V E S TI N G I N TH E D ATA E C O N O M Y

F U TU R E O F Q U AN T I TAT I V E I N V E S TI N G I N TH E D ATA E C O N O M Y

Extremely Confidential. Not for Public Distribution. | DATA CAPITAL MANAGEMENT | 23

Consistent approach to analyzing company value through its use of predictive modeling and rules-based execution. Real-time

adjustments to portfolio risks and strategy weightings are made in response to specific trading environments.

MACHINE-LEARNING ALGORITHMS

Dynamically adapt portfolio allocations

among mean reversion, momentum, and

event-driven strategies to capitalize on specific

market conditions

Combine advanced self-learning models and

proven risk management to diversify the

portfolio and produce superior uncorrelated

risk-adjusted returns

Model market context, company-specific inputs,

and risk factor analysis for rules-based portfolio

allocation among strategies

NEW DATA SOURCES

Breadth: Ability to identify non-obvious

correlation factors from novel data sources

including news, images, social networks,

macroeconomic feeds, etc. and link to price

movements in securities, currencies,

commodities, etc.

Depth: Analyzes, prioritizes and monitors big-

data input and its impact using systematic and

quantitative metrics

Speed: Real-time, rules-based extraction and

interpretation of information based on event

triggers from over 20,000 leading global

newswires, online newspapers, aggregators,

and blogs in under 5 seconds

CUTTING-EDGE TECHNOLOGY

Accommodate big data feeds and multi-factor

risk systems with fast, scalable infrastructure

and code

Employ sophisticated systems to codify

fundamental statistics for 3,000 US exchange-

traded companies and ETFs

Utilize advanced artificial intelligence algorithms

to predict idiosyncratic shocks, market

sentiment, and cycle duration

TAB L E O F C O N TE N TS

Extremely Confidential. Not for Public Distribution. | DATA CAPITAL MANAGEMENT | 24

DATA CAPITAL MANAGEMENT OVERVIEW

WHAT IS THE DATA ECONOMY?

WHAT NEW CHALLENGES DOES THE DATA ECONOMY CREATE FOR INVESTORS?

THE FUTURE OF QUANTITATIVE INVESTING IN THE DATA ECONOMY

LESSONS FROM THE INDUSTRIAL REVOLUTION

L E S S O N S F R O M TH E I N D U S TR I AL R E V O L U TI O N

Extremely Confidential. Not for Public Distribution. | DATA CAPITAL MANAGEMENT | 25

James Pierpont Morgan

Suggestions for FinTech start-ups in the Data Economy

• Standardize industry taxonomies

• Be the best at depth of information for a given vertical

• Take responsibility for the veracity of information

• Focus on speed of delivery

• Focus on historical consistency

• Make adjustments visible to upstream users

• Focus on permissible data use rights

• Don’t be all things to all people

Source: google images

L E S S O N S F R O M TH E I N D U S TR I AL R E V O L U TI O N

Extremely Confidential. Not for Public Distribution. | DATA CAPITAL MANAGEMENT | 26

Standardize Industry Taxonomies

Source: google images; Wikipedia

For illustrative purposes only. Back test results are not indicative of future returns.

An International Securities Identification Number

(ISIN) uniquely identifies a security. Its structure is

defined in ISO 6166. Securities for which ISINs are

issued include bonds, commercial paper, stocks

and warrants.

U.S. Railway System 1830 - 1850 Ford plant Dearborn Michigan; 1928

Extremely Confidential. Not for Public Distribution. | DATA CAPITAL MANAGEMENT | 27

0

20

40

60

80

100

120

140

160

$95

$100

$105

$110

$115

$120

$125

Millions

AAPL

Volume Close

AAPL Price and Volume 1/12/15 – 2/5/15

AAPL News Volume and Sentiment 1/26/15 – 1/29/15

1/26/15 1/28/15 1/29/151/27/15

Sentiment Score Alone is Not Sufficient

L E S S O N S F R O M TH E I N D U S TR I AL R E V O L U TI O N

Be the Best at Depth of information for a given vertical

Source: DCM Analytics’ Google Images

For illustrative purposes only. Back test results are not indicative of future returns.

• Sentiment score alone is insufficient for

unsupervised systematic execution:

• ~200 identified articles in universe

• ~40% neutral

• ~40% positive

• ~20% negative

James Pierpont Morgan

L E S S O N S F R O M TH E I N D U S TR I AL R E V O L U TI O N

Extremely Confidential. Not for Public Distribution. | DATA CAPITAL MANAGEMENT | 28

Take Responsibility for your Veracity

Source: Google Images

For illustrative purposes only. Back test results are not indicative of future returns.

Eads Bridge June 14th 1874Does your product do what you say it does?

Andrew Carnegie first used steel because he was in the bridge building business. Up until

the “elephant parade” of June 14th 1874, bridges were made of Iron and could not

successfully cross the Mississippi River and thus connect the East Coast to the West

Coast. Following this success, he transformed America and powered the Industrial

Revolution…. Not by bridges; but through steel

L E S S O N S F R O M TH E I N D U S TR I AL R E V O L U TI O N

Extremely Confidential. Not for Public Distribution. | DATA CAPITAL MANAGEMENT | 29

Suggestions for Start-ups in the Data Economy

• Standardize industry taxonomies

• Be the best at depth of information for a given vertical

• Take responsibility for the veracity of information

• Focus on speed of delivery

• Focus on historical consistency

• Make adjustments visible to upstream users

• Focus on permissible data use rights

• Don’t be all things to all people

Got API?

[email protected]

Our goal is to help drive a standard for other players in our ecosystem to coalesce around

• This approach helps avoid the “tragedy of the commons” and maximize collective ROI

• We are set up to be “early customers” of new technologies / beta releases

• We drive a culture of focused and targeted feedback

For those with self-interests congruent to ours; let’s help each other