US006349291B1

(12)

(54)

United States Patent

(10) Patent N0.:

Varma

(45) Date of Patent:

METHOD AND SYSTEM FOR ANALYSIS,

DISPLAY AND DISSEMINATION OF
FINANCIAL INFORMATION USING
RESAMPLED STATISTICAL METHODS

(73) Assignee: Attractor Holdings LLC, Greenwich,

CT (US)
Notice:

Subject to any disclaimer, the term of this

patent is extended or adjusted under 35

U.S.C. 154(b) by 0 days.

(21) Appl. No.: 09/489,364

Jan. 21, 2000
(22) Filed:
(51)

Int. Cl.7 .............................................. .. G06F 17/60

(52)

US. Cl. .......................................... .. 705/35; 705/37

(58)

Field of Search ............................ .. 705/35, 36, 37,

705/38

(Sep./Oct. 2000): pp. 5562.*

Gardner, G., and Stone, D. Estimating Currency Hedge
Rations for International Portfolios, Financial Analysts
Journal, vol. 51, No. 6 (Nov./Dec. 1995): start p. 6.*
Choong, N.K., and R. McLeod, Jr. Expert, linear models,
and nonlinear models of expert decision making in bank
ruptcy prediction: a lens model analysis, Journal of Man
agement Information Systems, vol. 16, No. 1, (Summer
1999): pp. 189206.*
Mitchell, M., and Stafford, E. Managerial decisions and
longterm stock price performance, The Journal of Busi
ness, vol. 73, No. 3 (Jul. 2000): pp. 287329.*
Laster, D. Measuring gains form international equity diver

si?cation: The boofstrap approach, journal of investing:

NeW York; vol. 7, No. 3, (Fall 1998): pp. 5260.*
(List continued on next page.)
Primary ExaminerVincent Miller
Assistant ExaminerDaniel S Felten

(74) Attorney, Agent, or FirmKenyon & Kenyon

(57)
ABSTRACT
The present invention provides a method and system for the

U.S. PATENT DOCUMENTS

4,541,115 A

5,283,856
5,359,699
5,598,439
5,774,878

* 2/1994 Gross et al.

* 10/1994 Tong et al.
* 1/1997 Wagner
* 6/1998 Marshall

A
A
A
A

9/1985

statistical analysis, display and dissemination of ?nancial

Werth ........................ .. 282/14

WWW. The present invention utilizes resampled statistical

methods for the analysis of ?nancial data. Resampled sta
tistical analysis provides a meaningful and reasonable sta
tistical description of ?nancial information, Which typically

. 348/13

escapes modeling using parametric methods (i.e. assump

5,790,177 A

8/1998 Kassatly

*
*
*

4/1999 White, Jr. .................... .. 705/1

6/1999 Maggioncalda et al. .... .. 705/36
8/1999 Nevo et al. ................. .. 705/36

6,003,018 A
6,012,043 A

* 12/1999 Michaud et al. ..

* 1/2000 Albright et al. ..
Makivic

.. ... .. ... .

705/36
705/36

5/2000

6,064,984 A

5/2000 Ferguson et al. ........... .. 705/36

FOREIGN PATENT DOCUMENTS

8-221573
8-221574
10-512985

* 8/1996
* 8/1996
* 12/1998

data over an information netWork such as the Internet and

395/51
395/22
375/326
705/35

5,893,069 A
5,918,217 A
5,946,666 A

6,064,662

JP

OTHER PUBLICATIONS

References Cited

(56)

JP

Feb. 19, 2002

Hansson, B., and Persson, M. Time Diversi?cation and

Estimation Risk, Financial Analysts Journal, vol. 56, No. 5

(75) Inventor: Samir Varma, Greenwich, CT (US)

(*)

US 6,349,291 B1

. . . ..

705/36

tions of a Gaussian distribution). The present invention

provides at least a GUI that provides functionality for user

input of statistical queries, a statistical computation engine

that performs statistical analysis of ?nancial data and a

graphical rendering engine that generates graphical display

of statistical distributions generated by the statistical com

putation engine. According to one embodiment, the present

invention employs a parallel processing architecture to
speed generation of the resampled statistics.

31 Claims, 17 Drawing Sheets

US 6,349,291 B1
Page 2

OTHER PUBLICATIONS

Kemp A. W. Bootstrap Methods: A Practitioners Guide,

Biometrics, vol. 56, No. 1 (Mar. 2000): Start p. 317.*
Caudill, S., and Holcombe, R. Speci?cation search and
levels of signi?cance in econometric models, Eastern
Econonic Journal, vol. 25, No. 3 (Summer 1999): pp.
289300.*

Deis, D., and Hill, R., An application of the bootstrap

method to the simultaneous equations model of the demand

and supply of audit services, Contemporary Accounting

Research, vol. 15, No. 1 (Spring 1998): pp. 8899.*
Magnar, L., and Steinar, E., Exact con?dence intervals
generated by conditional parametric bootstrapping, Journal
of Applied Statistics (May 1999): pp. 447459.*
Coleman, M., Lynford, L., and James, F., A neWly inte
grated and dynamic approach to securitiZed real estate risk
analysis and valuation Real estate Finance, vol. 12, No. 1

(Spring 1995): Start p. 68*

Putnam, B. Is the trend still my friend? Global Investor,

No. 66, (Oct. 1993): Start p. 21.*

Fiellin, D., and Feinstein, A. Bootstraps and jackknives:

NeW computerintensive statistical tools that require no

mathematical theories Journal of Investigative Medicine,

vol. 46, No. 2, (Feb. 1998): start p. 22*

BerkoWitZ, J., and Kilian, L., Recent Developments in

Bootstrapping Time Series, manuscript, Sep. 25, 1996.*
Michaud, R. The MarkoWitZ OptimiZation Enigma: Is Opti

miZed Optimal), Financial Analysts Journal, (Jan/Feb.

1989): pp. 3142.*
* cited by examiner

U.S. Patent

Feb. 19, 2002

Sheet 2 0f 17

QR\SEst

US 6,349,291 B1

ma(w\maJyms
=65
T
m
6

32
2:
C-

m5E22826-52

582m 5

E532 85

0 am:c22658.2

9E26-

0 0 5262 3:

U.S. Patent

Feb. 19, 2002

m8:Sow rm5?m-

Sheet 3 0f 17

US 6,349,291 B1

GRN

56:2:.

a:

U.S. Patent

Feb. 19, 2002

Sheet 4 0f 17

US 6,349,291 B1

305
Client ID - 310

Client Password - 315

Portfolio" - 320

Alert Rules* - 325

E-mail Address - 330

Billing Parameter - 335

Preference Parameter 1 - 340(1)

Preference Parameter 2 - 340(2)

Preference Parameter N-l - 340(N- 1)

Preference Parameter N- 340(N)

H6. 3

405
Investment ID - 410
Investment Name - 415

Investment Data* - 420

H6. 4

U.S. Patent

Feb. 19, 2002

Sheet 5 0f 17

US 6,349,291 B1

Rule ID - 510

Rule Object - 515

H6. 56

Investment ID - 520
Function - 525
Period - 530

Operator - 535
Percentile Value - 540

Sample Size - 545

Replications - 550

H6. 50

f507

U.S. Patent

Feb. 19, 2002

Sheet 6 0f 17

US 6,349,291 B1

605
Investment ID - 610

Return Object] - 625(1)

Return Object 2_- 625(2)

Return Object 3 - 625(3)
WW

Return Object N-1 - 625(N-1)

Return Object N - 625(N)

H6. 63

625
Date - 630
Value - 635

H6. 60

U.S. Patent

Feb. 19, 2002

Sheet 7 0f 17

US 6,349,291 B1

705

Function prototype ID - 710

Function Prototype Object - 715

H6. 7

805

Plot Type ID - 810

Plot Parameter - 825(1)

Plot Parameter 2 - 825(2)

Plot Parameter 3 - 825(3)
/\___/_\__/\/\/

Return Object N-1 - 625(N- 1)

Return Object N - 625(N)

FIG. 6

U.S. Patent

Feb. 19, 2002

Sheet 8 0f 17

US 6,349,291 B1

m5

mm6R

Sag

22@5986035698:23

2as,2%TH.i?

uESwz mno

AQRVQlRHLZ-ilNkR

56 : 25 0:E82

9%t9,

Q
16%
A
H

Q2E6Q5S3

60$5925m

U.S. Patent

Feb. 19, 2002

Sheet 9 0f 17

US 6,349,291 B1

Generate EDF.
PM?

A
6

____________________________________________________________ "

935

Bootstrap

/ / ;Rep|icat|0ns

_________________________________ __

Bootstrap

3 / Samples

U.S. Patent

Feb. 19, 2002

Sheet 10 0f 17

Re uest for Resampled

cSltatistical Analysls

Receive Input Parametersl~1010

r
Request Analysis
From @1015
RSA

Request Plot from GRE /\/1025

Transmit Plot Data Front @1029

End Subsystem
1030

H6. 70

US 6,349,291 B1

U.S. Patent

Feb. 19, 2002

Initiate Resampled

Sheet 11 of 17

1135

1105

Statistical Method

Bias Reguested

ll

Set Sample Space

US 6,349,291 B1

4/1115

yes

Perform Bias initialization /\/ll40

v

Calculate Memory

Requirements. Re uest
Memory Range in S ared
Memor Area and Output
emory Area

Algorithm

/\/lll7

Transfer Selected Sample @1119

S ace Data from Raw

Financial database to

Transform Data

"50

Portion of Shared Memory

Area

f1160

M = "TKO/N)

ll

Calculate Basic Statistics

%

"20

For Range of Lag lnteryals /\/ll25

Calculate Autocorrelations
(a) and Partial
Autocorrelations

ll
Initiate Parallel Processing @1165

l
Parallel
processin

70

Complete.

TV

Determine Lag Factor (N) @1130

That Minimizes
Autocorrelation (a) ;
Number of Consecutive

Notify Back End Processor @1175

and Transmit Memor

Periods of Data to PlCk at

Address in Output Da a

One ?me (N)

Area
V

_ Terminate

initiation Routine

H6. 7 7

"80

U.S. Patent

Feb. 19, 2002

Sheet 12 0f 17

Initiate Parallel

Processing

Perform Load Balancing /\/1210

Analysis
V

Request Parallel Process on @120

Processor 1-P

All

1230

Parallel Processes

"0

Comgleted
yes

Assemble Data Output Data @1240

Area

Notify Control Server of

Completion

@1250

Initiate Parallel

Processing

HG. 72

1260

US 6,349,291 B1

U.S. Patent

Feb. 19, 2002

Sheet 13 of 17

US 6,349,291 B1

( initiate Process ) V1305

V

Determine Output Address @1310

Range
it

Set Random Start Point in @1320

i360

Notify Sergegsgompleted
f Build Bootstrap Sample
x*replr'cation
v

K1335

Pick N Consecutive
Store in Temporary Array

Periods Using Bras and

r,

(1340

Count = Count + 1

1345
Count = M

yes

y
K1350 f Calculate Replication
Calculate s(x*"P"'""')
swephmon)
v

Store s(x*replication)

(1355

1355

U.S. Patent

Feb. 19, 2002

Sheet 14 0f 17

US 6,349,291 B1

initiate Bias )VMOS

i
Separate Sample Space Into /\/I41O

Up Day Set and Down Day

et

Pick Random Number r.

Between 0-1

@1420

1430

1440

Pick Up Day

Pick Down Day 1450

1460

@
H6. 74

U.S. Patent

Feb. 19, 2002

Sheet 15 0f 17

US 6,349,291 B1

U562 m

EQ?w?62:a52

$25

0?

.- $.1I-8

gQ??I0%$0$.00lkgo
2-1.123
w
W
N.
J.
mm
2
:
a
s
8
x
cm
MN2
w

.1Il.i I.1-1-

W
m
m05

,.:wmm-\I? :

i|Q\~w_
wW
82
i w
.2
w
o

m
F.

$3

m.-I

$0
,

5H2%2

U.S. Patent

Feb. 19, 2002

Sheet 16 0f 17

US 6,349,291 B1

B82%Q2 E62:52

E682351

mg522

$3

$2

$3
saouaunooo 1o luamad

a?

U.S. Patent

Feb. 19, 2002

Sheet 17 0f 17

US 6,349,291 B1

$6

NS8355am;Q2352G0B%:

as
E
6
2
5
8
2
1
2.062O25

@6$8E682S:5QE932 22W. 1@8G0B:zaegs ?afm

\
\
M

\a
V
1
.
=
.m:- \.:-

. :-

.L
Z
[It
_
F
_
_,
_
_
__

hOl. -U2meai?sm

a
m,

m.
V2:
\
m

m. m
mg wI

0Q
$08.080N2w;:3Q32.080ad02

$8@:2Em2632

US 6,349,291 B1
1

METHOD AND SYSTEM FOR ANALYSIS,

DISPLAY AND DISSEMINATION OF
FINANCIAL INFORMATION USING

statistical description of ?nancial information, Which typi

cally escapes modeling using parametric methods (i.e.,

assumptions of a Gaussian distribution).

RESAMPLED STATISTICAL METHODS

The present invention includes a ?nancial information

netWork node that is coupled to an information netWork such
as the Internet. The ?nancial information netWork node
includes a front end subsystem, a resampled statistical

FIELD OF THE INVENTION

The present invention relates to the area of electronic

analysis engine (RSAE) and a graphics rendering engine

information systems. In particular, the present invention

relates to a method and system for the delivery of ?nancial
information using resampled statistical methods over an
information netWork.

10

BACKGROUND INFORMATION

Investors and ?nancial analysts rely upon electronic infor

mation systems for the delivery of accurate ?nancial and
investment data and analysis in order to devise meaningful
investment strategies. The groWth of the Internet and World

15

(GRE). The front end subsystem provides a graphical user

interface (GUI) that alloWs clients also coupled to the
information netWork to submit requests for resampled sta
tistical analysis of various ?nancial investments and receive
graphical display of the results. The RSAE performs resa
mpled statistical analysis of ?nancial data in response to user
queries and incorporates routines to preserve temporal cor

relation in ?nancial data, Which necessarily provides more

accurate analysis. In addition, the RSAE provides for user
control of a number of parameters to simulate various

Wide Web (WWW) highlights the potential for global

?nancial environmental conditions. For eXample, according

distribution of real time or near real time ?nancial

to one embodiment, the RSAE alloWs a user to simulate

information and analysis. For example, a number of WWW

sites provide ?nancial information to clients such as inves

either bull or bear market conditions by setting a bias

parameter that controls a degree of randomness in the

tors and ?nancial analysts.

HoWever, conventional ?nancial information sites do not

resampling process. The GRE generates a graphical display

of statistical distributions generated by the RSAE.

provide meaningful analysis tools to accurately analyZe,

25

forecast and predict the behavior of ?nancial markets. Con

According to one embodiment, the present invention

employs a parallel processing architecture to speed genera

tion of the resampled statistics. The parallel architecture is
afforded by the nature of the resampling algorithm itself,

ventional technology for delivery of ?nancial information

over information netWorks such as the Internet typically
alloWs users to track returns for various investments and

Which permits the ?nancial data to be vectoriZed. This

perform rudimentary statistical analysis (e.g., computation

of the mean and standard deviation) for these investments.
HoWever, these rudimentary statistical functions are not

parallel processing architecture provides at least tWo sig

ni?cant advantages. First, the architecture permits the deliv
ery and processing of ?nancial data in compressed time

useful to investors in forecasting the behavior of ?nancial

markets because they rely upon assumptions that the under

statistical analysis. In addition, the parallel computation

lying probability distribution function (PDF) for the ?nan

frames, Which facilitates real time or near real time

35

scheme provides the ability to perform statistical analysis on

cial data folloWs a normal or Gaussian distribution, Which is

a large number of ?nancial entities (e.g., a mutual fund or

generally false.

hedge fund) through a Weighting process.

The true distribution of returns for any ?nancial market

(and thus of a trading strategy) is unknoWn. It is therefore

According to one embodiment of the present invention for

implementation on the Internet, a ?nancial information site

incorrect to rely upon a statistical model based on assump

is coupled to the Internet via a front end subsystem including

tions of normality (e.g., standard deviation). Typically, the

a WWW server. The ?nancial information site includes a

PDF for ?nancial market data is heavy tailed (i.e., the

histograms of ?nancial market data typically involve many

front end subsystem, a RSAE and a GRE. In addition, the

?nancial information site maintains a database of ?nancial
data for any number of ?nancial entities such as companies,
mutual funds etc. The ?nancial information site also main
tains a database of clients that have registered With the
?nancial information site and desire to obtain statistical

outliers containing important information). Thus, statistical

measures such as the standard deviation provide no mean

45

ingful insight into the distribution of ?nancial data.

Providing reasonable methods for the analysis of ?nancial

analysis of ?nancial data.

market data is essential for investors. Reasonable statistical

analysis of ?nancial data should at a minimum provide an
accurate assessment of potential ?nancial risk and reWard.

HoWever, conventional methods, Which rely upon assump

tions of a Gaussian distribution, are dangerous to investors
because these analyses understate the true risk and overstate
potential reWards for an investment or trading strategy.
Thus, this information is not generally useful and if relied

55

In order to perform a resampled statistical analysis, a

query is received from a client via the front end subsystem.
A client may specify a number of parameters including an
investment or investments (e.g., a portfolio) to be analyZed,
a ?nancial function, a sample siZe, a period, a type of plot
and a bias parameter, Which controls the randomness of the

resampling process. Based upon the parameters speci?ed by

the client, the RSAE performs a resampled statistical analy

upon promotes imprudent investment decisions. In general,

the heavy tailed nature of ?nancial data presents signi?cant

sis of relevant ?nancial data. The GRE then produces a

challenges in providing meaningful statistical analysis.

distribution plot based upon the output generated by the

RSAE, Which is presented to the client via the front end

SUMMARY OF THE INVENTION

subsystem.

The present invention provides a method and system for

the statistical analysis, display and dissemination of ?nan

According to one embodiment of the present invention,

the RSAE performs at least three types of ?nancial functions

cial data over an information netWork such as the Internet

and WWW. The present invention utiliZes resampled statis

tical methods for the analysis of ?nancial data. Resampled
statistical analysis provides a meaningful and reasonable

65

on ?nancial data. A gross rate of return function provides

analysis of the gross rate of returns for an investment over

a speci?ed time period. A maXimum draWdoWn function

provides analysis of a maXimum draWdoWn for an invest