Using R in Finance

Background About R and CRAN Finance Task View Selected Finance Packages Summary and Outlook

Using R in Finance

Dirk Eddelbuettel, Ph.D.
Debian and R Projects

Invited Panel Presentation at the
58th Midwest Finance Association Meetings
Chicago, IL, March 5, 2009

Dirk Eddelbuettel Using R in Finance


Outline

Background

About R and CRAN

Finance Task View

Selected Finance Packages

Summary and Outlook



Personal connections

Financial Econometrician by training, with twelve+ years as
a Quant in (Inv-)Banking, Hedge Funds, Prop Trading
Debian developer/contributor since 1995
Debian R co-maintainer since 1999, maintainer since 2001
R package author or co-author:
RQuantLib: R interface to the QuantLib libraries
digest: hash function ’digests’ of serialized R objects
random: R interface to true RNG via random.org
RDieHarder: R interface to DieHarder RNG testers
littler: R scripting front-end
RPostgreSQL: R interface to PostgreSQL RDBMS
Rcpp: C++ classes for extending R with C/C++ functions
RInside: C++ classes for embedding R in C++ programs



Personal connections

Other R Open Source activities:
R-SIG-Finance, R-SIG-Debian, R-SIG-HPC list ’owner’
Editor of CRAN Task Views Empirical Finance and High
Performance Computing
Mentor for two R-related projects (cran2deb; RPostgreSQL)
at Google Summer of Code
Co-organizer of the ﬁrst R in Finance conference in April
2009, Chicago (http://www.RinFinance.com)
useR! Tutorial lecturer on High Performance Computing
with R in 2008 and 2009, also at R / Finance
Other work-related R projects (spawning the CRAN
packages RLim, RBloomberg; others unreleased)
Quantian (live cdrom/dvd) author / developer
Associate Editor at Journal of Statistical Software



Where are we right now with R ?

R has grown tremendously in recent years, both in terms of
capabilities and users.
Widespread and increasing use in Finance.
S+ has been almost entirely eclipsed by R, much to the
chagrin of Insightful.
We now also have three companies offering commercial
support and extensions.
CRAN, the R package archive network, has grown
tremendously too: now north of 1600 packages.
There is more and more Finance content as shown in the
task view.



Illustration: Growth of r-help and r-sig-finance

q R−Help q
R−SIG−Finance
2000

q
q
q
q
Total R usage is difficult
q
to measure.
1000

q
Mean Number of Messages / Month

q Mailing list activity is
500

q sometimes used as a
proxy.
200

q
While r-help growth is
q slowing, several ’special
100

interest group’ lists such
q
50

as r-sig-finance
experience strong
growth.
20

1997 1999 2001 2003 2005 2007

Year

Source: Fox (2008, 2009), our calculations.


Illustration: Growth of CRAN packages
1.3

1.4

1.5

1.7
1.8

1.9
2.0

2.1

2.2
2.3

2.4
2.5

2.6
2.7

2.8
1700 q 1679
1400 q 1427
1200
q 1300
John Fox provided this
1000 q 1000

800
q 911 chart in an invited lecture
Number of CRAN Packages

q 739

600
q 647 at the last useR!
q 548
500 meetings.
400 q 406
q 357 Details, and more
300
q 273
metrics on R and the
q 219
200 dynamics of the R Core
q 162

q 129
group, are also in a
100
q 110 forthcoming R News
(soon: R Journal) article.
2001−06−21

2001−12−17

2002−06−12

2003−05−27
2003−11−16

2004−06−05
2004−10−12

2005−06−18

2005−12−16
2006−05−31

2006−12−12
2007−04−12

2007−11−16
2008−03−18

2008−10−20

Source: Fox (2008, 2009), our calculations


Aside: CRANberries

The CRANberries feed
summarizes both
changes to existing
packages, as well as
new packages.
Implemented in 200 lines
of R and using the tiny
blosxom blog engine, it
is available as a blog and
via an RSS feed.

Source:
http://dirk.eddelbuettel.com/cranberries/


CRAN and CRAN Task View Finance

Given 1600+ CRAN packages, navigation is difﬁcult.
Idea suggested a few years ago to have topical views.
These Task Views are not perfect, but the best currently
available approach at navigating CRAN packages.
We created the Empirical Finance view around 2004 which
contains several sections on
standard regression models
time series
ﬁnance
risk management
books / data sets
date and data management
Rather than repeating the task view, let us illustrate several
different packages.



Finance Task View



RQuantLib (QuantLib Group; Eddelbuettel)

option value option delta

1.0
80

QuantLib is a very

0.8
60

complete and very well

0.6
40

written C++ library for
0.4 quantitative ﬁnance.
20

0.2
0.0

RQuantLib, ﬁrst
0

50 100 150 50 100 150
option gamma option vega released in 2002,
40
0.04

provides access to a
30
0.03

subset of QuantLib’s
0.02

20

functionality directly from
R, in particular for
0.01

10

(equity) option pricing
0.00

0

50 100 150 50
Strike is 100, maturity 1 year, riskless rate 0.03
100 150
and some yield curve
Underlying price from 10 to 180
Volatility from 0.1 to 0.9 analysis.
> example(EuropeanOptionArrays)


Rmetrics (Wuertz et al)

Excess Distribution Tail of Underlying Distribution
1.0

qq q
qqq
qq
qq
qq
q q q q qq qq qqqq qqqqqqq
q qq q q qqq qqqqqq
q qqqqqqq
q qqq q qq
q qqqqq
q qqq
qq
qq
qq
qqqq
qqqq
qqq
qqq
qqq

5e−01
q qqq
qq
qq
qq
qq
q
q qq
qq
qq
qq
qq
qq
q
q qq
qq
qq
qq
q qq
qq
qq
q
qq
q
q q
q
q
q
q
q qq
q
q
q
q
q
q q
q
q
q
qq q
q
0.8

q
q
q qq
q
q
q q

Rmetrics, ﬁrst released
q
q q
q
q
q
q
q q
q
q
q
q
q
q q
q
q
qq qq
q
q
q
q qq
q
q

1−F(x) [log scale]
qq
q qq
q
q q

5e−02
q
q
q q
qq
q q
x [log Scale]

q q
0.6

q
q
q q
q
q
q
q q
q
q
q
q q
q
q

in 2004 is a
q
q
q qq
q
q
q q
q
q qq
q
q
q
q q
qq
q q
q
q
q
q qq
qq q
q
0.4

q
q
q q
q
q
qq q
q q

5e−03
q
qq q
q
q
q q

comprehensive set of R
q
qq q
qq
q q
q
q
q
u = 0.000599

qq
q
q q
q
q
q
0.2

q
qq q
q
qq
q
q
qq
q
qq q
qqq
q
q
qq
qq
qq
qq
q
q

packages.
q
qq
qq
q
qq
qq
5e−04
qq
qq
q q
qq q
qq q
0.0

q q q qq qq qqq
q qq q q q q

1e−03 1e−02 1e−01 1e+00 1e+01 1e−03 1e−02 1e−01 1e+00 1e+01

Fu(x−u) x [log scale] It covers time series,
Scatterplot of Residuals QQ−Plot of Residuals GARCH and volatility
modeling, Extreme Value
6

q q

q
q
q q
q
5

q q q
qq
q

Theory and Copulae,
6

q q
q q q
Exponential Quantiles

q
q qq q
q q q q
4

q
q q q q q
q
q
q qq
Residuals

q
qq
q q q q qq q qqq

derivative pricing,
q
q qq q qq
q q q q q q
q
q qq
q
4

q q q q q q qq q q qq
3

q qq q
qq q q q q
q q q qq q q q q q
qqq
q q q qq q qq qqq q qqq
q
qq q
q
qq qq q
q qq q
q
q qq
q
q
q
q
q q q qq q
q
q q q
q
q
q qq
q qq qq
q
q
q
q
q qq q q
q q
q q q q
q q q qq q qq
q
qq
q
q q q q q q q
q qq
2

q q qq q qq
qqq

portfolio analysis,
q
q qqq qq
q q qq q q qq
q
qq q q qq q qq q qq qq q q
q qq
qq
qq q q q
q q q qq
qq
qq
qq
q q q q q q q q qq
qq
qq
qq
q q q qqqqqqqqq q q q q qq q
q qq q q qq q qq q qq
2

q qq q q qq q q
q qq q q q q q qq q q qq
qq
qq
qq
q
q
q qq qq q q q
qq q q
q qq qq q q qqq qq
q q qq
qq
qq
qq
q q q q
q q
q qq qq q q q q q qq q
q
q q
q q q q qqq q q q q qqq q qq q q qq qq q q
qq q
q q qq
qq
q q
qq q q
qq q
qq q q q qq q q qq q qq q q q qq qqqq qq qq q q qqq q
q q q q qq
qq
qq
q
q
q q
qq q q q qq q q q qq
qq
1

q qq q q qq q q qq q q q q q q qqq q q q q qq q q qq
q q q qq
q
q q q
q q q q q qq q
q qq qq
qq
qq
qq
qq q q q q
q q q qq qq qq qq q q q qqqq q q q q q q
q q
q q q q qq q
qq q
q qq qq
qq
q
q
qq
q qq q q qqq
qq q qqq qqq qqqq qq qq qq q q q qq qqqqqqqqq q qqq qq
qq q qq
q q qq q q q q q q qq q qqqq qq qq qq q qqqq qq qq qq
qq qq q qq
qq
qq
qq
qq
qq
q qq qq q q q q q q qq

optimization and more.
qq q
q qq
q q qq qq q q qqq qq q q q qq
qq q
q q q q q qq q q q
qq qq
qq
qq
qq
qq
qq
qq q q qqqqq q q qq qqqq qqq qqq qqq qqqq qqq qqqqq qqq
q q qqq qq
qq qq q q q q q qq q q
qq
qq
q
qq q q q q q q qqq q q qq q q q qqq qq qq q qq q qq q q
q qqq q q qq q q
qqq qqqqqqqq qqqqqqqqqqq qqqqqqqqqqqq q qqqqqqq qqqq
qq q q qqq qq q qqqq qqqqq qqq q qq q q qq q qq qq qq
q q qq q qq q q qq q qq q qq
qq
qq
qq
qqq q q q q
q
q q q
q
qq q q q q q qqq q qq q qq qq q qq q qqqqqqq qq qqq qq q q
q qq q q q q q q q q q
q q q q q q q q q
q qq qq qq q qq
qq
qq
qq
qq
qq
0

0

0 200 400 600 800 1000 0 1 2 3 4 5 6

Ordering Ordered Data

> example(GpdModelling)


portfolio (Enos, Kane et al)

Cumulative Spread Return

20
Total return: −2.7
cap.usd
Total return: 28
smi
The portfolio
Return (%)

10
0 package provides code
−10
−20
Worst drawdown: −9 Worst drawdown: −− and data for real-life
Mar May Jul Sep Nov Mar May Jul Sep Nov
equity long/short portfolio
Cumulative Quantile Return analysis.
cap.usd smi The portfolioSim
30
quantile package adds simulation
high
Q9
Q8 support; the tradeCost
Return (%)

Q7
20 Q6
Q5
Q4
package analyses
Q3
10 Q2
low ex-post trade impact, and
0 the backtest package
Mar May Jul Sep Nov Mar May Jul Sep Nov permits to test portfolios.
Date

> example(backtest)


PerformanceAnalytics (Carl and Peterson)

Equity.Market.Neutral

This package provides a
99% VaR
99 % ModVaR

library of econometric
100

functions for
80

performance and risk
analysis of ﬁnancial
60
Density

portfolios.
It offers practioners and
40

researcher some of the
latest research in
20

analysis of both normal
0

and non-normal return
−0.010 −0.005 0.000 0.005 0.010 0.015 0.020 0.025
streams.
Returns

> example(chart.Histogram)


Summary

R is a very vibrant platform for data analysis, visualization
and programming with data.
Hence, R provides an excellent platform for academic
research and teaching — as well as investment research
and trading.
Finance has traditionally been one of the two key users of
the S language, and this constiuency has moved from S+
to R.
The number of dedicated Finance packages for R is
increasing as well, the R-Forge site is a good place to
watch (http://R-Forge.R-Project.org).



Outlook

If you don’t go with R now, you will someday.
– David Kane on r-sig-ﬁnance, 30 Nov 2004



R / Finance in April 2009



Contact

http://dirk.eddelbuettel.com

dirk@eddelbuettel.com


Using R in Finance

More Related Content

Viewers also liked (18)

Recently uploaded (20)

Using R in Finance