As the financial system becomes more complex, new methods to understand the inherent risks and dynamics are needed. Kimmo Soramäki will discuss how network analysis of large‐scale financial transaction data can be used to improve our understanding systemic risk. He will also show case studies how visual analytics and accurate data driven maps of asset correlations and tail risks can enable a stronger intuition of market dynamics.

1. Financial Cartography
February 6, 2013 – 12 p.m. U.S. Eastern Time
Kimmo Soramäki
Founder and CEO of
Financial Network Analytics (FNA)
• Audio: Use your microphone and speakers (VoIP) or call in using your telephone.
• Direct your questions to Staff via the Questions or Chat pane.
• To access this webinar audio via the internet, select “Mic & Speakers” under your Audio pane.
• Check that the audio on your computer is on and the volume is turned up.
• For technical assistance contact the Citrix webinar utility customer number: 1-888-259-8414
This material is the intellectual property of the presenter
and shall not be reproduced or used without the express written permission .

2. PRMIA Webinar
6 February 2013
Financial Cartography
Dr. Kimmo Soramäki
Founder and CEO
FNA, www.fna.fi

3. “When the crisis came, the serious limitations of existing
economic and financial models immediately became apparent.
[...]
As a policy-maker during the crisis, I found the available
models of limited help. In fact, I would go further: in the face of
the crisis, we felt abandoned by conventional tools.”
in a Speech by Jean-Claude Trichet, President of the
European Central Bank, Frankfurt, 18 November 2010
3

4. We did not have maps …
4

5. Eratosthenes' map of the known world5
c. 194 BC

6. … but what are maps
“A set of points, lines, and areas
all defined both by position with
reference to a coordinate system
and by their non-spatial
attributes”
Data is encoded as size, shape,
value, texture or pattern, color
and orientation of the points,
lines and areas – everything has
a meaning Political map of Europe
6

7. … but what are maps (contd.)
Cartographer selects only
the information that is
essential to fulfill the
purpose of the map
Maps reduce
multidimensional data into
a two dimensional space
that is better understood by
humans
Maps are intelligence
amplification, they aid in
decision making and build Map by John Snow showing the clusters of cholera
cases in the London epidemic of 1854
intuition
7

8. I. Mapping II. Mapping
Systemic Risk Financial Markets
8

9. Systemic risk ≠ systematic risk
News articles mentioning “systemic risk”, Source: trends.google.com
The risk that a system composed of many interacting
parts fails (due to a shock to some of its parts).
In Finance, the risk that a disturbance in the financial
system propagates and makes the system unable to
perform its function – i.e. allocate capital efficiently.
Not:
Domino effects, cascading failures, financial
interlinkages, … -> i.e. a process in the
financial network
9

10. Network Theory can be to Financial Maps
what Cartography is to Geographic Maps
Main premise of network theory:
Structure of links between nodes
matters
To understand the behavior of one
node, one must analyze the
behavior of nodes that may be
several links apart in the network
Topics: Centrality, Communities,
Layouts, Spreading and generation
processes, Path finding, etc.
10

11. Network aspect is an
unexplored dimension of data
Variables
Observations
11

12. First Maps Fedwire Interbank Payment
Network, Fall 2001
Around 8000 banks, 66 banks
comprise 75% of value,25 banks
completely connected
Similar to other socio-
technological networks
Soramäki, Bech, Beyeler, Glass and Arnold (2007), M. Boss, H. Elsinger, M. Summer, S. Thurner, The
Physica A, Vol. 379, pp 317-333. network topology of the interbank market, Santa
See: www.fna.fi/papers/physa2007sbagb.pdf Fe Institute Working Paper 03-
12
10-054, 2003.

13. More Maps: Federal Funds
1997 - 2006 Source: Bech, M.L. and Atalay, E. (2008), “The Topology of
the Federal Funds Market”. ECB Working Paper No. 986.
• 2600 loans worth $335
billion per day
• First Circle: 165
Second Circle: 271
Rest: 42
13

14. More Maps: Italian money market
Italian (very small)
Italian (small)
Italian (large)
Foreign
Source: Iori G, G de Masi, O Precup, G
Gabbi and G Caldarelli (2008): “A network
analysis of the Italian overnight money
market”, Journal of Economic Dynamics
and Control, vol. 32(1), pages 259-278 14

15. More Maps: DebtRank
August 2007 to April 2008 October 2008 to April 2010
Nodes: Financial institutions Source: Battiston et al, Nature
Links: Impact of an institution to another Scientific Reports 2-54, 2012
Nodes closer to center are more important (as are big and red) 15

16. Where are we today?
Regulatory response to recent financial crisis was to strengthen
macro-prudential supervision with mandates for more regulatory
data
“Big data” and “Complex Data”-> Providing tools and challenge to
understand, utilize and operationalize the data
Financial Networks are starting to get their own literature and
metrics different from other fields of Network Theory
16

17. Case: Oversight Monitor
The monitor will allow the
identification of systemically
important banks and evaluation of
the impact of bank failures on the
interbank payment system
(network is fictional)
Intraday Liquidity Network -example
The visualizations are available at www.fna.fi/webinar/prmia 17

18. Polling Question 1
Which types of networks are most important for
financial institutions and regulators?
1) Exposure/contagion networks
2) Trade/payment networks
3) Supply chain networks
4) Social networks
18

19. I. Mapping II. Mapping
Systemic Risk Financial Markets
19

20. Outline
Purpose of the maps
– Identify price driving themes and market
dynamics
– Reduce complexity
– Spot anomalies
– Build intuition
The maps: Heat Maps, Trees, Networks
and Sammon’s Projections
Based on asset correlations or tail
dependence
These methods are showcased for
visualizing markets around the collapse
of Lehman Brothers
20

21. Collapse of Lehman
Lehman was the fourth largest investment
bank in the US (behind Goldman Sachs,
Morgan Stanley, and Merrill Lynch) with
26.000 employees
At bankruptcy Lehman had $750 billion debt
and $639 billion assets
Collapse was due to losses in subprime
holdings and inability to find funding due to
extreme market conditions
Is seen as a divisive point in the 2007-2009
financial crisis
21

22. The Data
Pairwise correlations of return on 118
global assets in 4 asset classes
• Stock Exchange Indices
(e.g. Dow Jones)
• Foreign Exchange Rates
(e.g EUR/USD)
• Government Bonds
(e.g. Irish 10 year bond)
• Corportate Bonds
(e.g. EMU Corporate AAA, 1-3 years)
22

23. i) Heat Maps
January
Corporate 2007
Bonds
FX Rates
Government
Bond Yields
Correlation
-1
Stock
Exchange 0
Indices
+1
23

24. January 2007 t-2 t-1
Corporate
Bonds
FX Rates
Government
Bonds
Stocks
t+1 t+2
Corporate
Bonds
FX Rates
Government
Bonds
Stocks
24

25. ii) Asset Trees
Originally proposed by Rosario Mantegna in 1999
Used currently by some major financial institutions
for market analysis and portfolio optimization and
visualization
Methodology in a nutshell
1. Calculate (daily) asset returns
2. Calculate pairwise correlations of returns
3. Convert correlations to distances
4. Extract Minimum Spanning Tree (MST)
5. Visualize
25

26. Minimum Spanning Tree
A spanning tree of a graph is a subgraph that:
1. is a tree and
2. connects all the nodes together
Length of a tree is the sum of its links. Minimum spanning tree (MST) is a spanning
tree with shortest length.
MST reflects the hierarchical structure of the correlation matrix

27. Demo: Asset Trees
Color of node denotes asset class:
Dow Jones Size of node reflects volatility
(variance) of returns
Ireland 10 year Links between nodes reflect
government bond 'backbone' correlations
EMU Corporate
AAA, 1-3 years
- short link = high correlation
- long link = low correlation
EUR/USD
The visualizations are available at www.fna.fi/webinar/prmia 27

28. Correlation filtering PMFG
Balance between too much and too little
information, signal vs noise
One of many methods to create networks
from correlation/distance matrices
– PMFGs, Partial Correlation Networks,
Influence Networks, Granger Causality, Influence Network
NETS, etc.
New graph, information-theory, economics
& statistics -based models are being
actively developed
28

29. iii) NETS
• Network Estimation for Time-
Series
• Forthcoming paper by Barigozzi
and Brownlees
• Estimates an unknown network
structure from multivariate data
• Based on partial correlations
• Captures both comtemporenous
and serial dependence (partial
correlations and lead/lag effects)
29

30. iv) Sammon’s Projection
Proposed by John W. Sammon in IEEE Transactions on Computers 18: 401–409
(1969)
A nonlinear projection method to map a
high dimensional space onto a space of
lower dimensionality. Example:
Iris Setosa
Iris Versicolor
Iris Virginica
30

31. Demo: Sammon Projection
EMU Corporate
AAA, 1-3 years Color of node denotes asset class:
Dow Jones
Size of node reflects volatility
Ireland 10 year (variance) of returns
government bond
Distance between nodes reflects
EUR/USD similarity of correlation profiles
- close = similar
- far apart = different
31
The visualizations are available at www.fna.fi/webinar/prmia

32. Tail dependence
• Correlation is a linear dependence. The same visual maps can be extended
to non-linear dependences.
• Joint work with Firamis (Jochen Papenbrock) and RC Banken (Frank
Schmielewski), see www.extreme-value-theory.com
• Instead of correlation, links and positions measure similarity of distances to
tail losses
Tail Tree Tail Sammon
(Click here for interactive visualization) (click here for interactive visualization) 32

33. Polling Question
Where can financial data visualization provide
most value?
1. Data validation and exploration
2. Enhancing intuition
3. Add on to statistical analysis
4. Risk Management
5. Trading
33

34. “In the absence of clear guidance from existing analytical
frameworks, policy-makers had to place particular reliance on
our experience. Judgment and experience inevitably played a
key role.”
in a Speech by Jean-Claude Trichet, President of the
European Central Bank, Frankfurt, 18 November 2010
34

35. Blog, Library and Demos at www.fna.fi
Dr. Kimmo Soramäki
kimmo@soramaki.net
Twitter: soramaki

36. FINANCIAL NETWORKS – LESSONS
FOR RISK MANGEMENT
To learn more about this topic and attend a live
training course with Dr. Kimmo
Soramaki, please click on the following course
session for more information and how to
register.
NEW YORK
April 26, 2013
One-Day Training Course

37. Questions for the Presenter?
Send them in now by using your Question Pane in the webinar utility panel
Did you know that Sustaining Members attend thought leadership webinars at no additional cost?
Find out more about Sustaining Membership at http://prmia.org/index.php?page=membership
Find upcoming and recorded webinars at www.prmia.org/webinars
37

38. Thank you for attending this PRMIA Webinar!
Go to
www.prmia.org/webinars
to find a full schedule of
upcoming webinars
38