back start next


[start] [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21] [22] [23] [24] [25] [26] [27] [28] [29] [30] [31] [32] [33] [34] [35] [36] [37] [38] [39] [40] [41] [42] [43] [44] [45] [46] [47] [48] [49] [50] [51] [52] [53] [54] [55] [56] [57] [58] [59] [60] [61] [62] [63] [64] [65] [66] [67] [68] [69] [70] [71] [72] [73] [74] [75] [76] [77] [78] [79] [80] [81] [82] [83] [84] [85] [86] [87] [88] [89] [90] [91] [92] [93] [94] [95] [96] [97] [98] [99] [100] [101] [102] [103] [104] [105] [106] [107] [108] [109] [110] [111] [112] [113] [114] [115] [116] [117] [118] [119] [120] [121] [122] [123] [124] [125] [126] [127] [128] [129] [130] [131] [132] [133] [134] [135] [136] [137] [138] [139] [ 140 ] [141] [142] [143] [144] [145] [146] [147] [148] [149] [150]


140

method to attain further input reduction is to decorrelate the remaining input variables without throwing away any useful information they have to offer.

• Jurik Researchs DDR (www.jurikres.com), an add-in module for TradeStation, SuperCharts, and MS Excel.

Consumer Protection Agencies

Consumers have rights. Whenever you suspect you have been a victim of disreputable or illegal actions, consult the appropriate authorities. There are numerous agencies chartered to protect you and your investments, and their services are generally free. For a list of agencies accessible on the Web, go to the consumer protection section on the following Web page:

www.jurikres.com/cool/maincool.htm

For a short tutorial on how to recognize disreputable vendors of "super profitable" trading systems, go to this Web page:

www.jurikres.com/snake/main oil.htm

Test flight your trading System

Before placing real trades with your newly developed system, you may want to test your system with a simulated broker service. This way, you can gain experience placing trades over the phone, watch your trades perform in real time, and discover how fast slippage and commission expenses add up and eat into your profits. Here are some relevant Web pages:

www.auditrack.com

www.forex-trc.com

www.futuresnetwork.com

www.tradersnetwork.com

www.tradecomp.com

www.wallstreetstation.com

www.virtualstockexchange.com



Appendix D

Project Nimble: An Automated Equity Options Market Making System

Joseph Alotta

This is the account of an actual and profitable trading system and the ideas and events that came together in the summer of 1987 and through 1989. The persons mentioned are real people to whom the author owes a debt of gratitude for a very challenging and satisfying project experience.

Although some of the circumstances described here happened over a decade ago, the reader may draw on our collective experience for the implementation of new projects, which all share the same basic principles of having a real edge in the market and exploiting it in a way that minimizes the downside risk. Though the models have become more complex and the systems are faster and cheaper, trading is still about game theory and risk management.

About Steve fossett

I was introduced to Steve Fossett in March 1987. In addition to running a large trading organization, Steve had a number of sports exploits to his credit-swimming the English Channel, swimming the strait of Dardanelles, leading the Snowbird expedition to the top of Mount Everest, racing dogsleds in the Iditerod, running super marathons, skiing super marathons, holding the world record for the longest balloon flight (from Seoul, South Korea, to Vancouver, British Columbia), and most recently winning the Chicago to Mackinaw Island sailboat race, in a catamaran "Stars and Stripes," which won the Americas cup in Australia.

Fossett Corporation had about 350 market makers and about 20 staff persons. Some of the market makers traded with the firms capital and some used their own capital and were essentially clearing customers. The largest part of the market makers worked on the Chicago Board Options Exchange ( ) trading equity options, a



smaller part traded futures options on the Chicago Board of Trade (CBOT) or the Chicago Mercantile Exchange (MERC).

Steve Fossett has that determination, that super perseverance, that tough-it-out till you succeed attitude that is infectious; so when he asked me to build him a trading machine, I found myself saying "Yes" and then thinking about how to accomplish it. There existed one other machine in early development, the Timber Hill system, partially deployed at the time. Steve took me onto the floor of the to study this system, but all I could see were colored monitors and options prices, and they were not always on line. The inner workings of their operation were a highly proprietary secret.

THE SPX PIT

The pit designated by exchange authorities for automated systems was the SPX pit. This pit has European-style options on the S&P 500 index. The OEX was a much bigger and more active trading pit, containing American-style options on the S&P 100 index and I guess the motive was to increase the volume in the SPX pit, although it was still more active than all but the largest single stock option, IBM. Having European style options was an advantage to the systems designer; we could use the less computationally intensive Black-Scholes formula without having to worry about the binomial trees of the Cox-Ross-Rubenstein model.

SELECTION OF THE DATA FEED SERVICE

I quickly did some calling around, wanting to use the best datafeed available. Steve Schwartz had an option trading system that was popular upstairs at the Options Exchange called the Schwartz-a-tron. The market makers would eat their lunches and play backgammon to the yellowish glow of the Schwartz-a-tron.

We hired two team members in the first few days, Rob Davis Jr. and Pat Mac-Caulley. Together we analyzed alternative quote sources and we presented our findings to Steve Fossett. All vendors were expensive; the quality was difficult to determine, sometimes good, sometimes not.

This was the problem: S1AC was the only carrier able to get feeds directly from the exchanges; all other data vendors received their data from SIAC. Vendors would attempt to avoid problems of the feed during peak times, such as crashes, panics, opening rotation, and closing rotations, first by omitting option bids and offers, then by omitting underlying bids and offers and then finally by omitting option last sale prices. These were times we absolutely needed to be operating with full data.

We decided to go to SIAC ourselves directly, becoming our own data vendor. This was a very expensive prospect and Steve Fossett backed it 100 percent. We still did not have a design for the complete trading system, but we acted on faith, knowing that the best source of data would be crucial for its success. In the opinion of some of



[start] [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21] [22] [23] [24] [25] [26] [27] [28] [29] [30] [31] [32] [33] [34] [35] [36] [37] [38] [39] [40] [41] [42] [43] [44] [45] [46] [47] [48] [49] [50] [51] [52] [53] [54] [55] [56] [57] [58] [59] [60] [61] [62] [63] [64] [65] [66] [67] [68] [69] [70] [71] [72] [73] [74] [75] [76] [77] [78] [79] [80] [81] [82] [83] [84] [85] [86] [87] [88] [89] [90] [91] [92] [93] [94] [95] [96] [97] [98] [99] [100] [101] [102] [103] [104] [105] [106] [107] [108] [109] [110] [111] [112] [113] [114] [115] [116] [117] [118] [119] [120] [121] [122] [123] [124] [125] [126] [127] [128] [129] [130] [131] [132] [133] [134] [135] [136] [137] [138] [139] [ 140 ] [141] [142] [143] [144] [145] [146] [147] [148] [149] [150]