(12) United States Patent
`Kane
`
`(10) Patent N0.:
`(45) Date of Patent:
`
`US 6,317,728 B1
`Nov. 13, 2001
`
`US006317728B1
`
`(54) SECURITIES AND COMMODITIES
`TRADING SYSTEM
`
`(76)
`
`Inventor: Richard L. Kane, 5030 Champion
`Blvd., Suite 6-183, Boca Raton, FL
`(US) 33496
`
`( * ) Notice:
`
`Subject to any disclaimer, the term of this
`patent is extended or adjusted under 35
`U.S.C. 154(b) by 0 days.
`
`Omnitrader, “Investment analysis software 1996 Guide
`Computerized Trading Evaluation”, Future v25, n8, p37(1),
`Feb. 1991.*
`
`English D, How to choose and use investment software.
`(Compute’s Getting Started with Personal money Mange-
`ment) (Buyers Guide), Compute, V15, n4, pS12 (4), Apr.
`l993.*
`
`Gilliland S, “Take stock of your finances: investment soft-
`ware”, Computer Shopper, V14, n3, p512 (6), Mar. 1994.*
`
`* cited by examiner
`
`Primary Examiner—Vincent Millin
`Assistant Ex:1mz'ner—Hani M. Kazimi
`
`(74) Attorney, Agent, or Firm—Oltman, Flynn & Kubler
`
`(57)
`
`ABSTRACT
`
`In accordance with the invention there is provided a secu-
`rities trading system based on the principles of artificial
`intelligence. It includes a data acquisition system having an
`input communicating with a securities exchange for receiv-
`ing securities buy/sell data; a clock for generating clock
`times; a processing logic having inputs respectively com-
`municating with the data acquisition system and with the
`clock for assigning respective clock times to said buy/sell
`data; a decision logic having a repository for storing a set of
`buy/sell rules for buying and selling securities in response to
`the buy and sell data aligned with the clock times; and a buy
`and sell execution system having an input communicating
`with the decision logic for executing buy and sell orders in
`conformance with the buy/sell rules. In the securities trading
`system according to the invention,
`the decision logic
`includes at least one decision agent, the agent representing
`a respective buy/sell rule, wherein further the decision logic
`may include at least two decision agents, each decision agent
`representing a respective buy rule or a respective sell rule.
`Artificial intelligence is provided in that the decision agents
`are rewarded in a feed-back arrangement by being given
`added or reduced voting power when their recommendations
`are found to respectively result in successful or unsuccessful
`decisions. Thereby a self-learning feature is provided which
`results in improving the performance of the system as the
`number of transactions increase.
`
`9 Claims, 23 Drawing Sheets
`
`(21) Appl. N0.: 09/170,745
`
`(22) Filed:
`
`Oct. 13, 1998
`
`Int. Cl.7 .................................................... .. G06F 17/60
`(51)
`(52) U.S. C].
`............................................... .. 705/37; 705/36
`(58) Field of Search ................................ .. 705/35, 36, 37,
`705/38, 39; 235/379, 380; 340/825.26
`
`(56)
`
`References Cited
`U.S. PATENT DOCUMENTS
`
`...................... .. 705/37
`6/1987 Kalmus et al.
`4,674,044
`. . .. .
`. .. .. 705/37
`3/1992 Lupien et al.
`5,101,353
`
`3/1994 Guttcrrnail ct al.
`705/37
`5,297,031
`4/1994 Hartheimer et al.
`705/37
`5,305,200
`
`. .. .. 705,737
`12/1994 Togher et al.
`. . .. .
`5,375,055
`..................... .. 705/37
`3/1996 Hawkins et al.
`5,497,317
`l0/1996 Stolfo et al.
`........................... .. 705/8
`5,563,783 *
`9/1997 Cristofich et al.
`.
`. 705/37
`5,671,363
`5,845,266 * 12/1998 Lupien et al.
`. . .. .
`. .. .. 705/37
`5,873,071 *
`2/1999 Ferstenberg et al.
`705/37
`6,012,042 *
`1/2000 Black et al.
`705/36
`6,018,722 *
`1/2000 Ray et al.
`............................ .. 705/36
`FOREIGN PATENT DOCUMENTS
`
`
`
`8/1997 (EP) .
`*
`000790568Al
`1/1998 (JP) .
`*
`410011510/\
`WO009641293/351 12/1996 (VVO) .
`OTHER PUBLICATIONS
`
`Waters Information Services, Inc. “Citibank Tests FX AI
`System Compatible With F/X Trader?”FX Week, V1, n33,
`pN/A, Feb. 1991.*
`Schrnerken, Ivy, “Experts from the Promised Land to Wall
`Street”, Wall Street & Technology, V11, n13, p22(3), May
`1994*
`
`
`
`TS 1005
`
`1
`
`TS 1005
`
`

`
`U.S. Patent
`
`Nov. 13, 2001
`
`Sheet 1 of 23
`
`Us 6,317,728 B1
`
`Figure 1
`
`2
`
`

`
`U.S. Patent
`
`Nov. 13, 2001
`
`Sheet 2 of 23
`
`Us 6,317,728 B1
`
`
`
`Figure 2
`
`3
`
`

`
`U.S. Patent
`
`Nov. 13, 2001
`
`Sheet 3 of 23
`
`Us 6,317,728 B1
`
`BACKGROUND AND SUMMARY OF THE INVENTION
`
`Wealth Wizard ® Process Overview
`
`figure 1 presents an overview of the Wealth Wizard process.
`
`High Level View
`
`Figure 3
`
`4
`
`

`
`U.S. Patent
`
`Nov. 13, 2001
`
`Sheet 4 of 23
`
`Us 6,317,728 B1
`
`
`
`
`
`Q30;cacao...u..nu.B5_n2$
`
`.uZo.En.tJ3»8R..:3
`
`!a..u=£&.1u-8.3.:
`
`
`
`
`
`35.360.30030cunt£0920::us!EaE_£o-80¢
`
`
`
`
`
`5
`
`

`
`U.S. Patent
`
`Nov. 13, 2001
`
`Sheet 5 of 23
`
`Us 6,317,728 B1
`
`mT2:9“.
`
`
`
`.E::=..SUE.ii:88...
`
`
`
`
`
`E....._..__sI32...:IE...is.
`
`93.3..3.333-£85
`
`
`
`.—=9...auouuzm.._.__=.§.._
`
`_..%._O98...
`
`
`
`
`
`ammouam5.0:".2:3...
`
`-
`
`‘>
`
`ans)
`
`=a.a>n:1?can
`
`I...
`
`I
`
`,-
`
`is,
`
`:9?.2E....33I.39.:I.3.E.3=9.»3.233>..51!33..3E3.:-61.!
`
`
`
`$923:85...33.:3!:.\a.3=.33:.
`
`
`
`9....:9}B3.:9!9:.
`
`
`
`as.35..an.33.»as.32
`
`8»
`
`2_a!>
`
`-flmgfix!!.,_-
`
`
`.4.-.395$2!
`
`53>
`
`.39.1.3:
`
`53>.rs?B3,.
`
`LII.“-rt.
`
`8»
`
`.:._Ez3.
`
`uo.3._..§9
`
`95.3.3
`
`956.}
`
`ih_l.=IIE
`
`3..//l!:sa_?..i..i.r
`
`33.612395:
`
`54d_....l=a9a:D§B
`
`32
`
`E3
`
`nuooaml
`
`6
`
`
`
`
`
`
`

`
`U.S. Patent
`
`Nov. 13, 2001
`
`Sheet 6 of 23
`
`Us 6,317,728 B1
`
`
`
`a2
`
`c2.5:
`
`
`
`
`5ululainlz8...35;as.324..25..
`
`
`
`.Z.u.c§._3.3-.5?..o.B.._i93.2.39In
`
`
`
`
`
`23.4%.:.53....I.33:...5.525%.:33-3595:
`
`
`
`sum_flt..u.}.35?535?o..1a_..g9:uaoau.__D6.B
`
`.Ea.mw...au.sa:aEa_§=..€E....£IE.s!.a=.unaE/n§o£b.9.a9.E5n
`
`
`
`
`0I
`
`
`
`
`
`aaeuoamE9:a__.E§._
`
`
`
`caneroam.
`
`
`
`ammooam29.".2:5...
`
`
`
`
`
`A..~...”__N.€..-1-..._....-1....-..t~.w._,....5..-,_.,..;....\
`
`
`/..+a_87u1||:|!.!.o-..l»
`/...//\.
`.anI.l|uIll!-ll.
`
`.2,2..>%
`
`7
`
`

`
`U.S. Patent
`
`Nov. 13, 2001
`
`Sheet 7 of 23
`
`Us 6,317,728 B1
`
`.93.93
`
`3.81m
`
`.._SU<oz3oIi..u7
`I-.__..2
`
`
`
`
`mzaamooma_m=<:._§w_......_.§........_::..:.
`
`u—_U=._UCG_-fl:—fl>W.|.l.kin:5.3.82.3an.IF
`
`
`
`lad52¢ItI999:areain
`
`91!
`
`
`
`can5.3....:_n§.E
`
`—.3m
`
`3.
`
`8489...
`
`:21.‘@m\_E_.‘Ag:isIIii3!!
`
`flho.-tsDo£.IiI
`
`
`
`
`
`8
`
`

`
`U.S. Patent
`
`Nov. 13, 2001
`
`Sheet 8 of 23
`
`Us 6,317,728 B1
`
`
`Inlelllgant
`
`Agents|EVALUATEAGENTS]
`
`lsnunamdlmamsfir
`
`IaIdaIwlyds...lnin|m.ulo
`
`IHIBIOOL
`
`9
`
`

`
`U.S. Patent
`
`Nov. 13, 2001
`
`Sheet 9 of 23
`
`Us 6,317,728 B1
`
`
`
`2:._._3a._..m=._o:
`
`10
`
`10
`
`
`

`
`U.S. Patent
`
`Nov. 13, 2001
`
`Sheet 10 of 23
`
`Us 6,317,728 B1
`
`
`
`3|!3
`5;...c3..
`..__o=a..=i.._
`
`533
`
`
`
`22a:
`
`._o=._...§w.=3._Ea:
`
`_uaoa2.lu.__EIaWH
`
`
`
`.fi3un_._.3..._fiua.
`
`:__§..€a2._...h%
`
`.3.23u.8:
`
`__s.3383...as
`
`...&.._ii.
`
`
`
`
`
` «.818{L8.3..T$.u...¥:.nu...TLA1.Q3228._w1h.."_...:n!_...uvn1..flan..
`
`IE22.51..as...3-...
`
`
`
`
`
`
`
`
`..j..._..a._8....=.eE35...13T.......||Iu11.|i|||
`
`ii?
`
`..u_u9.a.eaa>.ouL
`951.._:=..l>.uu
`
`®
`
`WL
`
`11
`
`11
`
`
`
`

`
`U.S. Patent
`
`N
`
`32fl.011teehS
`
`US 6,317,728 B1
`
`
`
`
`
`£3.50_a=.s.._a:.=.:a=.o..:.:uU2.2.3..
`
`
`
`
`
`29:9”.
`
`.....e._o..3..W50502250
`
`§._.__...,2.312.3.33:51....:5./\.\083,0/N.
`a.2l|-2IHLU3.8»:
`
`:53.2,.
`
`12
`
`>®
`
`12
`
`

`
`U.S. Patent
`
`2
`
`32«I021tBCh
`
`US 6,317,728 B1
`
`mN_.2:9“.
`
`B..>V.838:0...-5o._o=..oa.N.550
`
`tsI8>|»I.5Vluz.50MAv;48%313hmL.
`8!:-3.9..13:55Iaicsnllao$2e32.c..._.4oz£
`
`S3
`
`
`
`acoaacuzat..:=...§..<ou>_u:<
`
`o._o=3..E2.53..§.m.__8.s..m
`
`
`
`
`13
`
`

`
`U.S. Patent
`
`Nov. 13, 2001
`
`Sheet 13 of 23
`
`Us 6,317,728 B1
`
`o__o_Eon_E9:2.5..o._m-._om3=m
`
`
`
`
`
`a:c=oao=..........=.o3<on>.a:<
`
`on?2:9“.
`
`
`
`—O.._Ou_._.¢O._mmz<.._
`
`14
`
`14
`
`
`

`
`U.S. Patent
`
`Nov. 13, 2001
`
`Sheet 14 of 23
`
`Us 6,317,728 B1
`
`
`
`
`
`==.Em4:5..a.§._aoust.9.3....
`
`m..m::m_u_
`
`aim
`
`finances.
`
`1!.3aim
`
`
`
`...>...o.£.§m.._am
`
`fiofimmzBo.
`
`15
`
`15
`
`

`
`U.S. Patent
`
`Nov. 13, 2001
`
`Sheet 15 of 23
`
`Us 6,317,728 B1
`
`§g.£§ .§P
`
`—
`
`16
`
`

`
`U.S. Patent
`
`Nov. 13, 2001
`
`Sheet 16 of 23
`
`Us 6,317,728 B1
`
`mp239“.
`
`
`
`._=....=mEflaoi
`
`17
`
`17
`
`

`
`U S Patent
`
`Nov. 13, 2001
`
`Sheet 17 of 23
`
`Us 6,317,723 B1
`
`99.5:
`
`
`.26:
`
`.33..9.2.9mamaa.=.=un_.u:_Euo._
`
`
`
`
`
`Eozzmsex...2,53.
`
`18
`
`

`
`U.S. Patent
`
`Nov. 13, 2001
`
`Sheet 18 of 23
`
`Us 6,317,728 B1
`
`
`
`«Egan.Ju3
`
`3..fi.._
`
`A3333.u\1--..\5a.w
`.,8
`
`..5...,:....z..
`
`J
`
`2..
`
`62.
`
`39.5:
`
`.
`
`
`.5:3359
`35.232$923.:as
`
`95.:-:5
`
`8:3:
`
`-.3.1:9}
`
`8»
`
`as:3-C
`
`22m23333....939
`
`
`.32.
`
`.8822.3N3....2=__um.a=_Euo._
`
`
`
`
`
`Ecfiu:9:25m...
`
`19
`
`19
`
`

`
`U.S. Patent
`
`Nov. 13, 2001
`
`Sheet 19 of 23
`
`Us 6,317,728 B1
`
`Figure 18
`
`20
`
`20
`
`

`
`U.S. Patent
`
`Nov. 13, 2001
`
`Sheet 20 of 23
`
`Us 6,317,728 B1
`
`:I'_q:un.eI NYSE MU D.»‘J.7»‘J.Bs'-.33
`
`WEflHflVHH$m|HIdMMhi a
`
`
`L-I
`
`‘;-Ha;IqCutd:———--------
`gar; aw nsmou II15s.
`__..___.wmm.____._.w_,““.3
`
`I
`
`;
`f
`
`L‘
`2
`
`[HIE-IEI—F.I "III!
`4*" EIIIIEEZI I
`I9
`"‘
`'
`“I1.".'lEI-' IIKT
`
`‘I11’
`
`W M
`
`I
`"
`
`‘PEP EIIJEIIEIIEEIEEUVEEEEI-EEEI
`IPFEIII
`IEEEIZEEIZEEEIEEIEEI
`|1IIEIIfl—III—IIl-EH I
`IS? LEIIIEIILE§I—iE]—E-
`EEIIEIIIEEIIHEEIIIEH-W .
`IEEIEEIEEEIEEEEEIII “M
`I
`“W EIII'.‘IE:1Il‘.-LTIICEI am
`I
`‘TIN EIIEEIEIEEEHEEEI-E3 am
`I
`“F EIUEEIEE I W
`WET I
`W EIIlfl§—§lI‘.?.ZI]1ll
`W—-1111---E |
`--ZZZZZII1—Xjj
`I
`—-in--i!:—-—- |
`
`-1111111-TIT
`
`m..fiéfi¥ém£@...“_WH
`
`
`
`21
`
`

`
`U.S. Patent
`
`Nov. 13, 2001
`
`Sheet 21 of 23
`
`Us 6,317,728 B1
`
`urauaruvu-MW. it H
`mfimfi
`u.1r1m!_.«.
` 2"}!
`iillliilllllllli
`II-IIrI,npn-II*II!IIIlItIIw24
`wuumwmwuuwmwmmm.
`
`__’-—LCjTj—1__ __ ___... .
`
`I113
`
`Mil
`
`2
`
`
`
`?
`
`i'J!E'i”.l!IIIIli'.' W
`2mmsawm
`l5§EVlCliil1 “IEfii"'F
`1IiI|lIIMIIII M
`inmgilll
`l|I|I||II||lI|
`I mnnn
`
`
`
`
`
`,5
`R
`
`I
`2!:
`*.:,
`;§
`9
`J }L!£.fi%iHi@ a.awa&1;a1_kg2;i% C{Ji3“§
`._
`12
`.m'”._
`b_
`I
`
`.. cm.:ms£1Ann.L|p
`
`Figure 20
`
`22
`
`22
`
`

`
`U.S. Patent
`
`Nov. 13, 2001
`
`Sheet 22 of 23
`
`US 6,317,728 B1
`
`Learning run on Kane25
`
`$20,000
`
`$15,000
`
`.
`
`$(15,000)
`
`r
`
`r
`
`r
`
`5 9131721252933374
`
`23
`
`+ After trade profit
`
`——...Pre-trade profit
`
`-+-Profit per trade
`
`—Linear (Pre-trade
`
`profit)
`
`—Linear (After trade
`
`profit)
`
`Figure 21
`
`23
`
`

`
`U.S. Patent
`
`Nov. 13, 2001
`
`Sheet 23 of 23
`
`US 6,317,728 B1
`
`Credit
`$117,250.00
`
`$38,750.00
`$54,562.50
`
`$48,375.00
`
`$54,562.50
`$77,312.50
`$48,250.00
`
`s54_,1a7.5o
`
`$82,562.50
`$18,500.00
`
`$524-.312.50
`
`$812.50
`$400.00
`52123
`$412.50
`
`Bought
`Sold Short
`Sold Short
`
`Bought
`Bought
`Bought
`Sold
`
`First day with updated agents
`Trans
`Time
`Order #
`Soid Short
`1 0:1 1 :34 RGJO001
`1 1:25:42 RGJ0002
`11 233:‘! 9 RGJOOO3
`12:1 9:11 RGJ0004
`12:25:06 RGJo005
`13:14:02 RGJ0006
`13:32:56 RGJ0007
`13:39:28 RGJ0008
`13:45:57 RGJ0009
`1 3:53:25 RGJ001 0
`14:45:06 RGJ001 1
`14:59:47 RGJ0012
`1 5:04:48 RGJ0013
`15:17:28 RGJ0014
`1 5:1 9:04 RGJ001 5
`1 5:23:18 RGJ001 6
`
`Bought
`Sold Short
`Sold Short
`Sold
`
`Bought
`Bought
`Bought
`Bought
`Sold shod
`
`15:3T:1 7 RGJ001 7
`15:47:09 FEGJ0018
`15:50:11 RGJ001 9
`15:54:02 RGJ0020
`
`Bought
`Sold
`Sold
`
`Shares
`
`Symboi
`1000 MRK
`1000 MRK
`1000 PEP
`1000 WMT
`1000 BA
`1000 PEP
`1000 WMT
`1000 BA
`moo BA
`1000 WMT
`1000 K0
`1000 BA
`1000 BA
`1000 KC
`moo wavrr
`1000 AOL
`1000 WMT
`1000 WMT
`1000 AOL
`1000 BA.
`
`Term
`Price
`117 1/4 Day
`117 1/8 Day
`38 3/4 Day
`54 9/16 Day
`48 5/8 _ Day
`3313/1a Day
`54 9/16 Day
`48 3/8 Day
`43 3/8 Day
`54-_ 9116 Day
`7‘? 5/16_Day
`48 1/4 Day
`48 114 Day
`77 1/4 Day
`54 1/8 Day
`82 1/3 Day
`54 _- 3116 Day
`544 1.14 Day
`82 9116 Day
`48 1/2 "Day
`
`shares A
`
`20000
`
`Figure 22
`
`24
`
`24
`
`

`
`US 6,317,728 B1
`
`1
`SECURITIES AND COMMODITIES
`TRADING SYSTEM
`
`The invention relates to a securities (the term “securities”
`is in the following to be understood to include “securities
`and/or commodities”) and/or commodities trading system
`that includes a computer arrangement communicating with
`a securities exchange, and has inputs for receiving buy and
`sell data. The computer arrangement is capable of evaluating
`the buy/sell data and issuing buy/sell orders in accordance
`with a plurality of buy/sell rules, i.e. “agents,” stored in the
`system. A feedback arrangement monitors the success and
`failure of the respective buy/sell agents and assigns rating
`powers, i.e. weightings, to the buy,/sell agents in order to
`implement a learning process for gradually improving the
`system performance based on past and continuously accu-
`mulating experience of the agents.
`
`BACKGROUND OF THE INVENTION
`
`A program-trading software system is disclosed to be
`available to individual investors, small trading houses and
`investment managers. The trading system monitors a port-
`folio of securities in real time, executing trades automati-
`cally. The system is targeted to Intra-Day trading, a method
`of trading well suited to computer automation. The system
`performs a kind of careful, tireless monitoring, aimed to give
`the user a daily return on his/her investment, while mini-
`mizing risk.
`The methods and formulas employed by the system were
`developed over a number of years of active trading, and take
`full advantage of 14 years of experience with pattern analy-
`sis and mathematical modeling.
`The system may increase assets by leveraging daily
`fluctuations aiming to beat the financial market, the market-
`makers, institutional investors and the brokerage houses.
`With the recent advent of deep discount electronic
`brokers,
`transaction fees have dropped to where active
`trading can be highly profitable. Large fund managers are
`limited by the size of the funds they manage, They cannot
`buy and sell large quantities of stock without impacting the
`price. On buying, the price goes up, and on selling it goes
`down.
`
`SUMMARY OF THE INVENTION
`
`The following is needed to practice the invention: soft-
`ware programs as disclosed herein; money to invest; an
`operating program such as Windows® 95 or NT, or the like;
`an internet connection; and a brokerage account.
`Investing can be risky, since it is impossible to predict
`stock market fluctuations or company events with any
`accuracy. Frequently stocks plummet in overnight trading,
`leaving an individual investor unprotected during that time.
`In general, bad news is remembered longer than good news.
`Abad quarterly report from a company may take months to
`recover from, while an item of good news may be forgotten
`in days. Generally, stock prices fluctuate 1—3% each day, and
`by leveraging these fluctuations as outlined herein one may
`realize a return on investment. A combination of real time
`
`monitoring a portfolio of stocks and no holding positions
`overnight (trading flat) can shield an investor from loss
`while maximizing gain.
`The invention embodies a process for trading securities
`that operates as follows: On a typical trading day a com-
`modity or a stock may take a small dip at e.g. 9:35 am, reach
`its high at 11:30 am and decline until 3 pm. perhaps with a
`
`10
`
`25
`
`30
`
`35
`
`40
`
`45
`
`60
`
`65
`
`2
`small run up at close of the day. The system buys, through
`a selected one of the agents, a stock or commodity at a
`morning dip and sells it at a mid-morning high. It then sells
`the stock or commodity short and buys to cover at a lower
`afternoon price. One makes money on the way up, and more
`on the way down. If possible, the system will end the day
`without holding any stock .
`.
`. but it if does, it will be short
`position. It is far more likely a stock will go down overnight
`than go up in value.
`By monitoring one’s stock holdings continuously during
`the day, and watching trends, the system will limit one’s
`losses by dumping the stock should things go awry, while
`maximizing profits by riding out the trends. If one holds a
`long position overnight, one may have uncontrolled risk.
`If the market is over-valued and headed for a correction,
`can one safely invest in this climate‘? The answer is yes! The
`system can help limit the risks.
`The system provides an extract file of trades, compatible
`with popular tax programs, to help prepare tax returns. This
`reporting can save time at tax time.
`Since it is difficult to watch and monitor a stock ticker
`during the day, a lapse in attention can be costly. Getting
`caught up emotionally may keep one from making good
`decisions. The system does not get tired, does not panic and
`does not get greedy, but always operates under the rules
`embedded in the agents.
`More importantly, brokerage firms do not let one place
`sell and stop orders on the same shares. That is to say, one
`can either try to protect oneself with stop loss orders, or one
`can try to profit with sell orders, but one cannot do both. The
`system effectively lets one do both by monitoring stocks
`continuously.
`The system enables a person to make money while on
`vacation. It does the drudge work of monitoring the market.
`It may be arranged to send alphanumeric trade reports to a
`user by cellular phone or pager.
`Use of margin can increase one’s earning power. With
`margin the system may deliver a return each month. But
`margin costs money. It is a loan against one’s current stock
`holdings, allowing one to buy more stock. If one holds the
`stock for days, weeks or months, one pays margin interest
`even if the stock price loses ground. By actively trading, one
`incurs a fraction of the margin costs, and may not incur any
`at all, since one is borrowing and returning the same day.
`The system will also let one hold stocks with pre-set buy,
`sell and dump prices, allowing one to automate transactions,
`get pager notifications and View continuously updated prices
`while supporting existing strategies.
`In accordance with the invention there is provided a
`securities trading system having a data acquisition system
`having an input communicating with a securities exchange
`for receiving securities buy/sell data; a clock for generating
`clock times; a processing logic having inputs respectively
`communicating with the data acquisition system and with
`the clock for assigning respective clock times to said buy/
`sell data; a decision logic having a repository for storing a
`set of buy,/sell rules for buying and selling securities in
`response to the buy and sell data aligned with the clock
`times; and a buy and sell execution system having an input
`communicating with the decision logic for executing buy
`and sell orders in conformance with the buy/sell rules.
`In the securities trading system according to the
`invention, the decision logic includes at least one decision
`agent,
`the agent representing a respective buy/sell rule,
`wherein further the decision logic may include at least two
`
`25
`
`25
`
`

`
`US 6,317,728 B1
`
`3
`decision agents, each decision agent representing a respec-
`tive buy rule or a respective sell rule.
`According to a further feature,
`the securities trading
`system provides that the sell rule is a short sell rule and the
`buy rule is a long buy rule, and the decision logic includes
`at least one agent being responsive to one of the buy/sell
`rules, that agent being operative for generating a buy/sell
`order in response to the buy/sell data conforming to the
`buy/sell rule.
`The securities trading system according to the invention
`may further include a plurality of agents, each agent oper-
`ating in response to a dedicated one of the buy/sell rules, and
`wherein each of the agents has a respective input
`for
`commonly receiving the buy/sell data.
`The securities trading system may further include a feed-
`back connection from the current assets memory to each of
`the agents for conveying a cumulative number of merit
`points to a respective agent having issued a sell order for a
`successful trade.
`
`The invention further includes a method for trading
`securities with a securities exchange, the method including
`a data acquisition system having an input communicating
`with at least one securities exchange for receiving buy/sell
`data; a clock for generating clock times; a processing logic
`having inputs respectively communicating with the data
`acquisition system and with the clock for assigning respec-
`tive clock times to the buy/sell data; a decision logic
`including a repository for storing a plurality of buy/sell rules
`for buying and selling securities in response to the buy/sell
`data; the decision logic having a plurality of agents, each
`operating in response to a respective buy/sell rule for
`generating buy/sell orders for securities in conformance with
`the buy/sell data; the agents having outputs communicating
`with the securities exchange for executing the buy/sell
`orders; wherein the method includes the steps:
`(a) issuing to all agents a tentative buy short/sell long
`order for a given security;
`(b) soliciting from all agents a tentative buy short decision
`for the given security;
`(c) affirming with the decision logic the buy short decision
`if a majority of the agents have indicated an affirmative
`buy short decision; and
`(d) executing with an executing logic the aflirmed buy
`short order.
`
`The method may further provide the features of:
`(a) monitoring for a given length of time the security
`bought on the buy short order; and
`(b) when the security has accrued value sulficiently to at
`least cover the short buy plus a given profit, issuing a
`buy long order for the security.
`The method may further include setting the given length
`of time to end before the trading day of the buy short order,
`and/or monitoring for another given length of time with the
`decision logic the rates of success and failure of each agent
`and feeding back to each agent a cumulative merit quotient
`increment according to the cumulative rate of success and/or
`failure for the respective agent.
`Further objects and advantages of this invention will be
`apparent from the following detailed description of a pres-
`ently preferred embodiment, shown schematically in the
`accompanying drawings.
`BRIEF DESCRIPTION OF THE FIGURES
`
`FIG. 1 is a block diagram of the major components of the
`invention, showing hardware and software components and
`their mutual lines of interaction;
`
`10
`
`15
`
`25
`
`30
`
`35
`
`40
`
`45
`
`60
`
`65
`
`4
`FIG. 2 is a simplified flow chart showing the major steps
`performed in practicing the invention;
`FIG. 3 is a high level View of the flow of information and
`the major steps of a typical transaction;
`FIG. 4 is a block diagram of a typical embodiment of the
`invention showing its major function blocks and the process
`information flow between the blocks;
`FIG. 5 is a flow chart showing the major function blocks
`and the information and decision steps in performing a
`learning process for a successful long order transaction;
`FIG. 6 is a flow chart showing the major function blocks
`of a learning from success process as in FIG. 5, for a short
`order transaction;
`FIG. 7 is a flow chart for an evaluation procedure showing
`the major steps of the evaluation procedure with artificial
`intelligence and evaluation criteria;
`FIG. 8 is a flow chart showing the major steps in an
`evaluating procedure for evaluating the agents;
`FIG. 9 is a flow chart showing the major steps in deter-
`mining hold/sell decisions;
`FIG. 10 is a flow chart showing the major steps in
`performing a hold/sell evaluation procedure;
`FIG. 11 is a flow chart showing the major steps of
`portfolio capture, including handling of partial orders;
`FIG. 12::
`is a flow chart showing the major steps in
`processing buy,/sell-short data from the portfolio, including
`analysis of account transactions;
`FIG. 12b is a flow chart showing the major steps in
`analyzing buy/sell-short data from the portfolio, including
`analysis of account transaction, and parsing of the portfolio;
`FIG. 13 is a [low chart showing the major steps in parsing
`a trade message from a broker, including getting a message
`relating to sell/buy-to-cover;
`FIG. 14 is a flow chart showing the major steps in the
`timing loop for a typical transaction;
`FIG. 15 is a flow chart showing the major steps of a
`program startup procedure, and the major decision steps of
`the procedure;
`FIG. 16 is a flow chart showing the major steps in learning
`from a failure process related to long order flow;
`FIG. 17 is a flow chart showing the major steps in learning
`from a failure process related to short order flow;
`FIG. 18 is a printout of sample forms/executions, showing
`a trade for 900 shares short of e.g. stock TRV, wherein the
`trade record shows a sample trade;
`FIG. 19 shows sample form executions, similar to FIG.
`18, but with two trades in progress;
`FIG. 20 shows ample graph executions continued from
`FIGS. 18 and 19, visualizing system calculations; and
`FIG. 21 shows in graph form continual improvement Via
`the system’s self-learning feature
`FIG. 22 shows a trading pattern of the system on an
`average trading day.
`Before explaining the disclosed embodiment of the
`present invention in detail it is to be understood that the
`invention is not limited in its application to the details of the
`particular arrangement shown, since the invention is capable
`of other embodiments. Also, the terminology used herein is
`for the purpose of description and not of limitation.
`DESCRIPTION OF THE PREFERRED
`EMBODIMENT
`
`FIG. 1 is a block diagram of the major components of the
`invention, showing hardware and software components and
`
`26
`
`26
`
`

`
`US 6,317,728 B1
`
`5
`lines of interaction; wherein an executing
`their mutual
`device 11 is connected via respective “buy long” and “sell
`short” data channels 12,13 to a decision logic 14 composed
`of a plurality of “agents” 16.
`Each agent exists as a module or section of computer
`logic, physically stored in a computer memory 17, which is
`connected with and controlled by a central processing unit
`CPU 18 in conventional manner. Each agent 16 performs a
`respective buy or sell decision based on buy and sell rules
`embedded in each agent. All agents represent dilferent buy
`and sell rules, and all receive continuously the movements
`of the securities market(s) in general as received from the
`conventional ticker tape data as continuously being issued as
`a data stream from the various securities and/or commodities
`markets, being received on a data line 17 connected via
`conventional data transmission facilities from securities
`and/or commodities markets.
`
`The ticker-tape data as they arrive in continuous streams
`of data are extracted by the ticker-tape monitor 18 which
`retains only data that pertain to specific securities and/or
`commodities stored in the system’s Current Assets Memory
`19, connected via data line 21 to the ticker-tape monitor 18.
`All pertinent ticker-tape data are stored in an organized order
`in the Tape Cycle Store 22, and periodically examined under
`control of a Clock 23.
`
`A Data Monitor and Record Change Unit 24 keeps a
`running record of all periodically recorded data from the
`Tape Cycle Store via data line 26, and processes the data in
`accordance with a certain set of general rules pertaining to
`all securities and for commodities in inventory such as, for
`example, running averages for certain selected time periods,
`certain selected trends, such as price trends that are rising
`and falling in accordance with preset criteria, and so forth.
`All such selected prices, trends and moves of securities
`and commodities prices are maintained and kept updated in
`the data monitor and record change unit 24. The decision
`logic 14 is composed of a plurality of e.g. 160 intelligent
`agents, divided into, e.g. 80 long agents and 80 short agents,
`which may collectively issue “buy/sell” suggestions for
`securities transactions as they may pertain to one security
`transaction at a time. The decision as to the acquisition
`and/or disposal of positions in the securities and/or com-
`modities is made by each agent according to the rules
`embedded in each agent. Generally, all agents make a
`recommendation as to the disposition of a respective secu-
`rity and/or commodity and a vote is taken of all decisions of
`the respective agents by a voting algorithm contained e.g. in
`the decision logic 16. The result of the vote is transmitted via
`one of the “buy long” data channel 12 or the “sell short” data
`channel 13, and the decision is executed in the executing
`device 11, which transmits the corresponding order via an
`appropriate data channel, such as an internet data connection
`27, stock exchange data connection 28, or a data channel 29
`to a broker. A special data connection 31 may be provided
`to a so-called “super broker” who handles several types of
`trades.
`
`10
`
`25
`
`30
`
`35
`
`40
`
`45
`
`As a result of the execution of each trade, corresponding
`adjustment in val11e takes place for a respective security or
`commodity in the current assets memory 19. The adjustment
`may be positive as a result of a successful transaction or
`negative for an unsuccessful transaction. Each adjustment is
`fed back from the current assets memory 19 as a score via
`data channel 32 to be accumulated in each agent in a merit
`memory dedicated to each respective agent. In each agent,
`its voting power is weighted by the accumulated score. As
`a result, agents accumulating higher scores attain over time
`
`60
`
`65
`
`6
`increased “voting power” so that agents providing better
`decisions gain increased influence on the overall system’s
`performance. In other words, the system undergoes a “learn-
`ing process.”
`FIG. 2 is a simplified flow chart showing three major steps
`performed in practicing the invention, beginning with an
`“Intent to Purchase” 101, representing a pending buy or
`pending short transaction, which is followed by an “Acqui-
`sition” 102, which is in turn followed by an “Intent to Sell”
`103, followed again by “Intent to Purchase” 101, and so
`forth;
`FIG. 3 is a high level View of the [low of information and
`the major steps of a typical transaction;
`FIG. 4 is a block diagram of a typical embodiment of the
`invention showing its major function blocks and the process
`information flow between the blocks;
`FIG. 5 is a flow chart showing the major function blocks
`and the information and decision steps in performing a
`learning process for a successful long order transaction;
`FIG. 6 is a flow chart showing the major function blocks
`of a learning from success as in FIG. 5, for a short order
`transaction;
`FIG. 7 is a [low chart for an evaluation procedure showing
`the major steps of the evaluation procedure with artificial
`intelligence and evaluation criteria;
`FIG. 8 is a flow chart showing the major steps in an
`evaluating procedure for evaluating the agents;
`FIG. 9 is a flow chart showing the major steps in deter-
`mining hold/sell decisions;
`FIG. 10 is a flow chart showing the major steps in
`performing a hold/sell procedure;
`FIG. 11 is a flow chart showing the major steps of
`portfolio capture, including handling of partial orders;
`FIG. 12a is a flow chart showing the major steps in
`processing buy/sell-short data from the portfolio, including
`analysis of account transactions;
`FIG. 12b is a flow chart showing the major steps in
`analyzing buy/sell-short data from the portfolio, including
`analysis of account transaction, and parsing of the portfolio;
`FIG. 13 is a flow chart showing the major steps in parsing
`a trade message from a broker, including getting a message
`relating to sell/buy-to-cover;
`FIG. 14 is a flow chart showing the major steps in the
`timing loop for a typical transaction;
`FIG. 15 is a flow chart showing the major steps of a
`program startup procedure, and, and the major decision steps
`of the procedure;
`FIG. 16 is a flow chart showing the major steps in learning
`from failure process related to long order flow;
`FIG. 17 is a flow chart showing the major steps in learning
`from failure process related to short order flow;
`In FIG. 18, this trade record shows a sample trade on the
`system.
`FIG. 19 shows a sample forms/executions, similar to FIG.
`18, but with two trades in progress. 700 shares of K0 are
`held LONG for 73 minutes, while 1000 shares of PFE are
`held long for 41 minutes. The unrealized profit
`in the
`portfolio is #728. The position in K0 is slightly adverse but
`improving, while the position in PFE is $813 positive.
`Fourteen trades have been conducted with a net of $7788.15.
`In this run, 29 of the 30 stocks being tracked are open with
`15 up from previous close, 14 down from previous, 13 up
`from open and 15 down from open. The blue highlight on the
`Minutes:73 tag for KO indicates a pending sale of the
`security, due to the insullicient movement in the security.
`
`27
`
`27
`
`

`
`US 6,317,728 B1
`
`7
`FIG. 20 shows a sample forms,/execution to visualize
`system calculation and to illustrate the 2—3% daily variance
`of the securities selected by the progam to trade in. The
`vertical lines indicate the first decision points by the pro-
`gram. Note the high of 2.5% over the open price and the low
`of 0.3% of the open. Each block is a 30 minute portion of
`time from market open.
`FIG. 3 High Level View
`(1) Intelligent Agents
`The 200 intelligent agents currently in the system contain
`rules and logic which evaluate market and specific equity
`behaviors. Many of these agents are further controlled by
`system parameters. When an agent ‘fires’, the value associ-
`ated with that agent is added to the long or short ‘votes’ for
`the direction the equity will move. Agent values are con-
`tinually updated by the program as part of the learning
`process.
`(2) Certainty
`The system incorporates an additional check, after the
`intelligent agents have voted, that checks the certainty of the
`vote for each equity. Essentially the system asks itself, are
`you certain? This check is unique in that the inventor has
`found no reference to it in A.I.
`literature, and it greatly
`improves the trading accuracy of the system. This check
`looks