Markit Data Guide

12301 Research Blvd. Building IV, Suite 410 Austin, TX 78759 U.S. Help Desk Phone: +1-800-546-9646 (or direct +1-512-697-3000), select ext. 3400 U.K. Help Desk Free Phone: 0800 032 6063 Europe Help Desk Phone: +44 20 7190 2947 Help Desk Email: support@lim.com +1-512-697-3001 (Fax)
Part Number: 077_106 Date: July 24, 2007

Copyright 2007 by Logical Information Machines, Inc. Patented May, 1995 U.S. Patent No. 08/392, 612 All rights reserved. Markit is a registered trademark of Markit Group Limited. No part of this publication may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, recording, or otherwise, without the prior written permission of Logical Information Machines, Inc. Restricted Rights Legend Use, duplication, or disclosure by the Government is subject to restrictions as set forth in subdivision (c) (1) (ii) of the Rights in Technical Data and Computer Software clause at 252.227-7013. Logical Information Machines, Inc. 120 North LaSalle Street Suite 2150 Chicago, IL 60602 Phone: +1 (312) 456-3000 Product names mentioned herein are for identification purposes only and may be trademarks and/or registered trademarks of their respective companies. While every precaution has been taken in the preparation of this manual, we assume no responsibility for errors or omissions. Neither is any liability assumed for damages resulting from the use of the information contained herein. Logical Information Machines, Inc. may revise this publication from time to time without notice.

MARKIT DATA GUIDE

Table of Contents

iii

Table of Contents
Chapter 1: Introduction .................................................................................................................. 1 Credit Default Swaps ................................................................................................................ 1 Standard Credit Events (according to the ISDA 2003 agreement) ........................................... 2 Trading Idea Research ....................................................................................................... 3 Chapter 2: Credit Default Swap (CDS) Data in the MIM ................................................................ 5 Path, MIM Symbols, Columns ................................................................................................... 5 Consolidated Redcode ....................................................................................................... 5 Separate Redcode .............................................................................................................. 6 Consolidated GII Ticker .................................................................................................... 8 Consolidated CUSIP ......................................................................................................... 8 Separate CUSIP ................................................................................................................ 9 Ticker .............................................................................................................................. 9 Universal Alias ........................................................................................................................ 10 Examples ............................................................................................................................... 11

iv

MARKIT DATA GUIDE

Table of Contents

MARKIT DATA GUIDE

Chapter 1: Introduction

CHAPTER 1

Introduction
Credit Default Swaps
Investors in corporate debt instruments have long had to rely on credit agencies such as Moody's and Standard & Poor's to assess the level of risk associated with the debt obligations. These agencies use teams of analysts to analyze the company from a fundamental point of view, looking at the amount of debt, the balance sheet of the company and its business prospects to determine a credit rating. A revolution in credit-risk assessment is taking place with the development and growth of an over-thecounter market for a credit derivative known as a Credit Default Swap (CDS). A CDS allows debt holders to buy protection against the risk of debt default of the debt issuer from counter-parties who are willing to take on that risk in exchange for a periodic premium payment. The risk premium that the sellers of protection are willing to take in order to take on the risk of default for the debt issuer can be used as a market-based credit rating for that entity. If the credit default swap rate rises it is an indication that the risk of default of the debt issuer is rising. Since this market is an over-the-counter market with no central exchange and no standardized instruments, it has been difficult to track this risk premium with any degree of certainty because the deals are done directly between the purchaser and the seller of the risk protection. However a London-based company named Markit has become the de-facto collection point for this data. They have arranged deals with the major investment houses doing these sorts of deals whereby information about the deals is sent to Markit, and then Markit distributes the aggregate, anonymized data back to all the contributors. Now, for the first time, through a partnership between Markit and Logical Information Machines, Inc. historical data on Credit Default Swaps will be made available in a time-series format in LIM's Market Information Machine (MIM) suitable for analysis alongside other data sets such as debt-issuer price histories. The histories of the credit default swap rates can be plotted over time, or spread with those of other instruments. Movements in the stock prices of the debt issuers can be correlated against movements in the credit default risk premium in order to find profitable trades in either based on this analysis. The MIM/Markit data set includes Credit Default Swap data for some ~1800 equities which trade on U.S. exchanges, as well as equities that trade on international bourses. The data set includes rates on swaps denominated in all available currencies, for various protection terms and restructuring clauses. The 5 year spread is the benchmark term series because it is the most commonly traded and most liquid of the credit default swap contracts. As a result, the data will be more even, consistent, and easier to extract pertinent information where other spreads may have gaps. If for any reason a person wanted to view a

MARKIT DATA GUIDE

Chapter 1: Introduction

different spread, they are available in the MIM. Other Spreads available in the MIM include 6 month and 1, 2, 3, 5, 7, 10, 15, 20, and 30 year. CDS confirmations also specify the credit events, most notably bankruptcy, that will trigger a credit event and give rise to payment obligations by the protection seller and delivery obligations by the protection buyer.

Standard Credit Events (according to the ISDA 2003 agreement)

Bankruptcy Failure to pay affecting a minimum of $1 million Restructuring repayment affecting a minimum of $10 million or in ASIAN and EM markets Obligation acceleration Obligation default Repudiation/Moratorium

Restructuring is the most controversial of all these credit events. Therefore special specifications have been made to CDS contracts. These stipulations are necessary in avoiding mismatches in protection because short based assets tend to have higher recovery rates than long based assets. There are four different types of customized restructuring agreements, and as should be expected, they are all coded into the MIM separately. After a Modified Modified Restructuring event (MM) there must be a delivery of an asset with less than 30 months remaining. This type of restructuring event is most commonly used in Europe. In the MIM, this contract is coded as MM at the end of the column name as can be seen in this example. After a Modified Restructuring credit event, the delivery of an asset with less than 60 months remaining must follow. This type of restructuring is most common in the US, and in the MIM this is coded as MR. After a Restructuring Credit event, there is no time stipulation as the delivery of an asset of any duration must follow. This is most common in Asia, and is coded in MIM as CR. No Restructuring occurs when restructuring is not a credit event at all. This is coded in MIM as XR. If you would like to graph all these spreads side by side, a query in the MIM, such as this one, makes it quite simple:
LET @ = GII.F.NYSE SHOW MR: Spread05y_SNRFOR_USD_MR CR: Spread05y_SNRFOR_USD_CR MM: Spread05y_SNRFOR_USD_MM XR: Spread05y_SNRFOR_USD_XR WHEN Date is within 1 year

of of of of

@ @ @ @

MARKIT DATA GUIDE

Chapter 1: Introduction

Trading Idea Research

Thus far, we have not created any historical trading ideas for trading the actual credit default swaps; however, movements of the credit default swap have been successfully used as events which show promise in predicting future movements of the underlying companys share price. In most cases that have been effectively used so far, an event in the underlying equity, in conjunction with a particular event in the CDS (most commonly a percentage move), have yielded the best results as far as obtaining historically rooted trading ideas for the underlying equity. As far as point and click events that are CDS specific there are a few theoretical ideas that have not yet been sufficiently tested: When different types of Spreads move in different directions. A day in which the Modified Restructuring 5 year spread becomes smaller, the same 5 year spread with No Restructuring becomes wider (i.e., the same spreads with different restructuring rules, move in different directions). The different time frames of the spreads move in different directions (e.g., the 5 year spread widens while the 7 year or 1 year spread tightens). Combination moves combining the % move in the CDS contract, and the underlying equity.

MARKIT DATA GUIDE

Chapter 1: Introduction

Chapter 2: Credit Default Swap (CDS) Data in the MIM

MARKIT DATA GUIDE

CHAPTER 2

Credit Default Swap (CDS) Data in the MIM

CDS data from Markit is first loaded by a primary key of REDCODE, a Markit unique identifier for an issuing corporate or government entity, plus any other secondary identifiers (redcode+Tier+Ccy+DocClause) to make a unique key. If there exists a tie to a core equity and it can be identified and linked by a CUSIP provided by both Markit for the Redcode, and by Global Insight for the equity prices, then, LIM will have an alias ticker in the form GII.ticker.exchange (e.g., GII.IBM.NYSE ) which contains a certain consolidated subset of columns identified by Spread_ColumnName+Tier+Ccy+DocClause. Due to the extensive of number combinations, only those with data are persisted to the MIM.

Path, MIM Symbols, Columns

Data is provided in the MIM in 6 different value added reference entity formats in the following paths:
TopRelation:Credit_Derivatives:Markit

Data columns are also provided: Separately and completely by:

MIM Symbol+Tier+CCy+DocClause

Spread Columns are Consolidated by:

MIM Symbol with Spread_ColumnName+Tier+Ccy+DocClause

Consolidated Redcode
Markit CDS consolidated columns are defined by:
MIM Symbol="MKT.Redcode"

MARKIT DATA GUIDE

Chapter 2: Credit Default Swap (CDS) Data in the MIM

where only Spread_xx columns are consolidated. For the following examples, MKT.001ABC is AT&T Corp. Columns for the consolidated MKT.001ABC look like this:

Separate Redcode
Markit CDS columns are defined separately by:
MIM Symbol="MKT.Redcode"+Tier+Ccy+DocClause

Thus AT&T Corp's many combinations of available swaps are:

Chapter 2: Credit Default Swap (CDS) Data in the MIM

MARKIT DATA GUIDE

Columns for the separate MKT.001ABC_SNFOR_AUD_CR look like this:

MARKIT DATA GUIDE

Chapter 2: Credit Default Swap (CDS) Data in the MIM

Consolidated GII Ticker

Columns for the consolidated GII Ticker follow the form:
MIM Symbol = GII.ticker.exchange

Consolidated CUSIP
Columns for the consolidated CUSIP follow the form:
MIM Symbol = CUSIP.cusip+tier+ccy=docclause

Chapter 2: Credit Default Swap (CDS) Data in the MIM

MARKIT DATA GUIDE

Separate CUSIP
Columns for the separated CUSIP follow the form:
MIM Symbol = CUSIP.cusip,

since Markit provides the CUSIPs.

Ticker
Ticker columns follow the form:
MIM Symbol = TICKER.ticker.exchange

10

MARKIT DATA GUIDE

Chapter 2: Credit Default Swap (CDS) Data in the MIM

Universal Alias
CUSIP.cusip & TICKER.ticker.exchange are new universal aliases which can be used in any context when columns of data from different vendors are provided in independent MIM databases allowing for simple identifier use.
SHOW 1: Close of CUSIP.459200 2: Spread06m_SNRFOR_USD_MR of CUSIP.459200

where:
Close of CUSIP.459200

references the default core price from the GII data and references data from Markit.

Spread06m_SNRFOR_USD_MR

A LET statement can also be used:

LET theSec = CUSIP.459200 SHOW 1: Close of theSec 2: Spread06m_SNRFOR_USD_MR of theSec

Chapter 2: Credit Default Swap (CDS) Data in the MIM

MARKIT DATA GUIDE

11

Examples
The following shows three XMIM query examples that use the CDS data from Markit: Example 1: Weyerhaeuser Co closed down a 'big' -1.0% on Friday after its credit default swap rate spread fell by -10.5% on Thursday. Also on Thursday, WY had a high of $84.28 which was above the 52-week highest high. WY closed Friday with a share price of $82.25, but should we expect it to go higher in the near future based on the modified modified credit default swaps huge loss on Thursday. Question: How has WY performed when, in the last year, its credit default swap rate spread loses at least -4%?
SHOW CDS: Spread06m_SNRFOR_USD_MR of GII.WY.NYSE t+3: percent_move from today to 3 values later of WY t+4: percent_move from today to 4 values later of WY t+6: percent_move from today to 6 values later of WY WHEN Date is after 12/12/2006 AND Date is before 2/23/2007 AND 1 value percent_move of Spread06m_SNRFOR_USD_MR of GII.WY.NYSE less than -4

is

Answer: According to the 13 previous occurrences of this event, WY has shown a strong bullish edge that peaks 6 trading days after the event. WY rallies in 92% of the cases (12 of 13) by an average of 3.2% relative to the close on the event date. The average of the 1 decline is -0.1%. The overall return of the 13 cases is 2.9%. Four days from the event WY rallies in all 14 cases by an average of 2.46%.
Date 12/13/2006 12/15/2006 12/21/2006 01/02/2007 01/08/2007 01/09/2007 01/10/2007 01/12/2007 01/18/2007 01/19/2007 01/23/2007 01/29/2007 02/01/2007 02/08/2007 02/12/2007 02/14/2007 02/15/2007 02/22/2007 Day Wed Fri Thu Tue Mon Tue Wed Fri Thu Fri Tue Mon Thu Thu Mon Wed Thu Thu CDS 0.0534 0.0605 0.0555 0.0519 0.0495 0.0454 0.0434 0.0424 0.0538 0.0465 0.0321 0.0325 0.0436 0.0567 0.0595 0.0553 0.0503 0.0441 t+3 7.0324 -3.0863 1.3523 NaN 0.3880 1.6713 2.1438 -1.3562 3.6939 0.9697 0.6428 0.3319 4.4991 2.4813 5.0357 1.5571 1.4322 2.0224 t+4 6.8506 -2.3356 0.7829 NaN 1.1501 1.5320 1.0649 1.7123 4.1105 0.7138 0.8705 1.5932 4.8829 3.2536 5.8516 2.4644 0.8253 3.4188 t+6 6.4717 -1.7656 3.2028 NaN -0.0554 0.2925 4.0353 2.2877 4.3605 1.4411 0.4419 4.8460 3.4008 5.1146 6.5400 0.8460 4.4059 3.7438

Avg AvgPos AvgNeg PctPos

0.0487 0.0487 NaN 100.0000

1.8124 2.3503 -2.2212 88.2353

2.2789 2.5673 -2.3356 94.1176

2.9182 3.4287 -0.9105 88.2353

12

MARKIT DATA GUIDE

Chapter 2: Credit Default Swap (CDS) Data in the MIM

PctNeg Maximum Minimum StdDev ZStat Variance

0.0000 0.0605 0.0321 0.0082 5.9515 0.0001

11.7647 7.0324 -3.0863 2.3688 0.7651 5.6110

5.8824 6.8506 -2.3356 2.2397 1.0175 5.0162

11.7647 6.5400 -1.7656 2.3914 1.2203 5.7189

18 Occurrences

Example 2: Home Depot has closed down 8 consecutive days with 4 'large' five day declines. The most recent move of -0.4% resulted in a large five day decline of -1.6%. HD had a bearish reversal on Friday, the same day that its modified restructured credit default rate spread increased by 16.67%. Also as a result of the resent down moves, HD is trading below its 20 day moving average, but above its 200, 100, and 50 day moving averages. Question: What has happened HD in the past year, when S&P 500 and HD close down, resulting in HD trading below its 20-day moving average on a day in which its modified restructuring credit default swap rate spread increases by more than 1%?
SHOW CDS: Spread06m_SNRFOR_USD_MR of GII.HD.NYSE t+3: percent_move from today to 3 values later of HD t+42: percent_move from today to 42 days later of HD WHEN Date is before 2/24/2007 AND Date is after 4/1/2006 AND 1 value percent_move of SPX is less than 0 AND 1 value percent_move of Spread06m_SNRFOR_USD_MR of GII.HD.NYSE more than 1 AND HD is less than 20 value average of HD

is

Answer: According to the 10 previous occurrences of this event, HD has shown a strong bearish edge that peaks 42 trading days after the event. HD declines in 90% of the cases (9 of 10) by an average of -9.1% relative to the close on the event date. The average of the 1 rally is 4.7%. The overall return of the 10 cases is -7.7%. It should be noted that there is an occurrence of this event which has not peaked yet, which appears will finish higher than the close on the event date. This effectively makes the edge 82% correct (9 of 11) 42 days from the event; however, three days from the event, HD declines in 91% of the cases (10 of 11) which does take this most recent occurrence into account.

Date 04/21/2006 04/25/2006 05/03/2006 05/12/2006

Day Fri Tue Wed Fri

CDS 0.0164 0.0165 0.0199 0.0220

t+3 -0.5213 -0.0751 1.5853 -4.9750

t+42 -9.5829 -8.7087 -11.3451 -14.0500

Chapter 2: Credit Default Swap (CDS) Data in the MIM

MARKIT DATA GUIDE

13

05/22/2006 06/05/2006 06/07/2006 06/23/2006 06/30/2006 12/22/2006 02/23/2007

Mon Mon Wed Fri Fri Fri Fri

0.0266 0.0257 0.0273 0.0245 0.0282 0.0304 0.0273

-0.2338 -0.5606 -1.2258 -0.6866 -0.8941 2.1599 -3.3447

-10.7560 -8.9429 -5.4481 -6.4543 -4.7499 6.2998 -4.0283

Avg AvgPos AvgNeg PctPos PctNeg Maximum Minimum StdDev ZStat Variance

0.0241 0.0241 NaN 100.0000 0.0000 0.0304 0.0164 0.0048 5.0660 0.0000

-0.7974 1.8726 -1.3908 18.1818 81.8182 2.1599 -4.9750 1.9885 -0.4010 3.9541

-7.0697 6.2998 -8.4066 9.0909 90.9091 6.2998 -14.0500 5.3804 -1.3140 28.9488

11 Occurrences

Example 3: Walt Disney Company gapped down -1.2% resulting in a large five day decline of -2.7% on Wednesday, the same day that its modified restructuring Credit Default Spread increased. However, DIS mounted a bit of a comeback on Thursday with a 0.49% rally to finish the trading session with a share price of $34.39. Is there anything we can extrapolate from these recent moves based on the stock's history? Question: How has DIS performed in the last year when its 5-year credit default rate spread increases on a day in which DIS experiences a 'large' five day decline, followed by a rally on the next trading day?
SHOW CDS: Spread05y_SNRFOR_USD_MR of GII.DIS.NYSE t+1: percent_move from today to 1 day later of DIS t+3: percent_move from today to 3 days later of DIS t+4: percent_move from today to 4 days later of DIS t+5: percent_move from today to 5 days later of DIS WHEN Date is before 4/1/2007 AND Date is after 3/28/2006 AND 1 value percent_move of DIS is more than 0 AND 1 value ago 5 value percent_move of DIS is less than (30 day average of 5 value percent_move of DIS (1 * 30 day std_dev of 1 value percent_move of DIS ) ) AND 1 value ago 1 value percent_move of Spread05y_SNRFOR_USD_MR of GII.DIS.NYSE more than 0

is

14

MARKIT DATA GUIDE

Chapter 2: Credit Default Swap (CDS) Data in the MIM

Answer: According to the 8 previous occurrences of this event, omitting repeat occurrences within 5 trading days, DIS has shown a very strong bullish edge that peaks 4 trading days after the event. DIS rallies in 100% of the cases (8 of 8) by an average of 1.2% relative to the close on the event date.
Date 03/29/2006 05/23/2006 06/14/2006 07/19/2006 08/10/2006 11/15/2006 11/28/2006 12/18/2006 01/08/2007 01/29/2007 02/28/2007 03/14/2007 03/29/2007 Day Wed Tue Wed Wed Thu Wed Tue Mon Mon Mon Wed Wed Thu CDS 0.2247 0.2012 0.2084 0.1864 0.1958 0.1258 0.1267 0.1168 0.1168 0.1017 0.1036 0.0948 0.1404 t+1 0.0359 -0.2977 1.7428 -0.4785 -0.5071 1.1318 -0.0304 1.0438 -0.1449 1.0381 0.4088 0.1779 0.1163 t+3 0.2872 0.9262 1.3245 0.3759 0.2028 1.3154 0.6081 0.1450 1.4203 0.9515 -1.5474 1.0673 1.5411 t+4 0.1795 0.9262 1.8822 0.7519 0.4395 1.6519 1.6722 -0.9568 2.0580 1.4417 0.3504 1.9271 0.7270 t+5 -0.2513 -0.0992 1.5336 0.2734 1.0818 0.9177 3.9830 -0.9568 2.0580 1.6724 0.5255 4.2692 1.5121

Avg AvgPos AvgNeg PctPos PctNeg Maximum Minimum StdDev ZStat Variance

0.1495 0.1495 NaN 100.0000 0.0000 0.2247 0.0948 0.0465 3.2112 0.0022

0.3259 0.7119 -0.2917 61.5385 38.4615 1.7428 -0.5071 0.7021 0.4642 0.4929

0.6629 0.8471 -1.5474 92.3077 7.6923 1.5411 -1.5474 0.8224 0.8060 0.6764

1.0039 1.1673 -0.9568 92.3077 7.6923 2.0580 -0.9568 0.8760 1.1460 0.7674

1.2707 1.7827 -0.4358 76.9231 23.0769 4.2692 -0.9568 1.5308 0.8301 2.3434

13 Occurrences