Ken Deen's THE AGGRESSIVE TRADER(tm) "Seeking double-digit gains in one to four months" STATUS REPORT 1339 Virginia Rd. Vol. 1, No. 81 Montecito, CA 93108 August 26, 1992 (805) 565-2039 CompuServe: 72020,2050 This month, I present the new improved DESI Trading System(tm), as promised in the last Status Report, Vol.1 #70. The improvements, based on research I conducted in June and July of this year, dramatically enhance the System's historical performance. First, however, let us review my performance for the month of July and take a look at current market conditions. July 1992: Overview July was an excellent month for The Aggressive Trader(tm). The Model Portfolio was up 3.9%, and the Stock+Options Account was up 7.0%. However, I am still down for the year, with the Model Portfolio showing a loss of 5.9% since the first of the year, and the Stock+Options Account showing a loss of 11.2% since its inception at the end of March. See Tables 3 and 5 for details. By comparison, the NASDAQ Composite was up 3.1% in July and up 0.9% for the year to date. The S&P 500 was up 3.9% for July and up 1.7% for the year. So I have outperformed the market in July, but underperformed for the year. Comparing my results with mutual funds which follow strategies similar to mine, my year-to-date results start to look much better. Pacific Horizon Aggressive Growth was down 16% year-to-date on July 31, and Twentieth Century Ultra was down 13%. I am outperforming both of these aggressive growth funds. Both of these funds are rated A+ by Investor's Business Daily based on their very impressive 3-year track record. Market Commentary I continue to be bullish on growth stocks intermediate-term, but I am now cautious short-term. I expect a strong rally later this year, beginning sometime between now and late November, led by growth stocks. The recent August selloff was impressive, but not as impressive, volume-wise, as the powerful two-day rally we had on July 28 and 29. Short-term, the market must deal with the crisis in the dollar and the resultant backing up of long-term interest rates. This, and uncertainty over the election, may continue to weigh heavily on stocks in coming weeks. However, many bullish factors remain in place; see Vol.1 #80 for details. On August 25, the NASDAQ Composite closed at 554.22. This is very close to the low for the year, set on June 26, of 547.84. If this index were to set a new low, especially if such a break were to come on high volume, then I would have to re-think my bullish stance. Introducing an Improved DESI Trading System(tm) In this issue, I will share with you some of the highlights of the evolution of the DESI Trading System(tm). This represents a progress report on more than a year's worth of full-time investigation into trading systems based on earnings surprises. For those of you who are just interested in the bottom line, I will repeat here some of the statistics I presented last month in a "sneak preview". In historical back-testing from Nov. 16, 1990 through July 22, 1992, this improved DESI Trading System(tm) has produced 22 completed trades on paper, 19 of which were winners. All trades are 3 to 3.5 months in duration. Of the 19 winners, the average gain was 24.3%, versus an average loss for the 3 losers of 7.2%. The best gain was 108%, versus the worst loss of 14.8%. Commissions are ignored in these statistics, but the bid/ask spread and slippage are more than taken into account(1). In an effort to evaluate the trading system from an even more revealing angle, I am currently writing a computer program to track a hypothetical $25,000 investment using the DESI Trading System(tm) exclusively(2). I will report these results as they become available. I have utilized my 17-year expertise in software engineering to design, execute, and test this System. The Theory Behind the DESI Trading System(tm) It is well-known that positive earnings surprises can provide excellent trading opportunities. Most traders attempt to anticipate these surprises by buying the stock ahead of the earnings news. This results in a runup in the stock price a week or two prior to the announcement. In such a situation, when the earnings are actually announced, there may be little reaction; indeed, there may actually be a drop in the stock price, as traders "buy on the rumor and sell on the news". When this happens, the business press may refer to the earnings report as an "earnings surprise", meaning that the numbers are better than the expectations of analysts; but it is not a surprise to traders, since they were buying ahead of the news. On the other hand, when the earnings news is a genuine surprise -- to traders as well as to analysts -- it can spark a rally that can last for weeks or months. These are the surprises which the DESI Trading System(tm) detects. Factors such as the long-term growth record of the company are evaluated by the DESI Trading System(tm), but the most important factors are the strength of the earnings report and the strength of the market reaction to that report. Services do exist (I/B/E/S and Zacks) to track the expectations of analysts regarding upcoming corporate earnings, but no service exists to track the expectations of money managers, institutional investors, and traders -- the people who actually buy and sell large blocks of stock. In the absence of such a service, the market action of the stock itself provides some strong clues. Using these clues, the DESI Trading System(tm) infers which earnings reports came as the biggest surprise to these key market players. When one of these true earnings surprises hits the news wires, some very fast traders will buy the stock on the day of the news. This results in a strong up day on strong volume. Many times, some profit-taking takes place the following day and the stock may give back part of its gain. However, within a few days of the surprise, a second wave of buyers will come in, buyers who needed time to evaluate the earnings report and to research the company, and also buyers who waited for this profit-taking to run its course. This second wave of buying often receives additional stimulus from analysts who upgrade their rating on the stock and raise their estimates of future earnings. Value-oriented investors may also be among the early buyers, because after a very strong earnings report, trailing earnings jump up, causing the P/E to drop. Within a couple of weeks, the stock has risen in price sufficiently to attract the attention of momentum players -- those investors who look for stocks with the best relative strength. As an upward trend in the stock price becomes clear to the army of chart watchers and technical analysts, more and more momentum players jump in. This can create a snowball effect, as the stock has by now established a clear upward trend. One of the great beauties of this System is that it identifies stocks which attract investors of various stripes. Fundamental analysts are attracted to the stock because of the excellent earnings growth and improved outlook for future earnings; technical analysts like it because of the pickup in price, volume, and relative strength; and value-oriented investors like the drop in P/E. The Problem with Trading Systems When evaluating a trading system, it is very important to find out what historical data was available to the system's author during the system's development and fine-tuning. I recommend ignoring the performance of the system over time periods that were known when the system was designed or revised. Yes, that's right, I suggest ignoring the "19 winners and 3 losers" and other statistics I have cited in order to attract your attention. I will show you below how I arrived at these stunning statistics, but, in the final analysis, these statistics don't count. Anyone can design a trading system which looks great on known historical data. What counts is how the system performs over time periods which were unknown to its author. Obviously, the most important test of a trading system is how well it performs in actual practice, in real-time, under live market conditions. In my opinion, it is equally valid to test the system under historical market conditions, over time periods which were not available to the author. I will use the phrase "data that count" to refer to performance data over time periods before and/or after those used in the system's design. The DESI Trading System(tm) in its current form was designed using buy candidates from Nov. 16, 1990 through July 22, 1992. This is not a long period of time, but it does include a spectrum of market conditions, from a roaring bull market to a severe correction of bear market proportions. There are many other trading systems which have been written about in books, magazine articles, newspapers, and advertisements, and they all sport impressive track records. However, I have yet to see any which make this important distinction between historical data that were used in the design of the system versus "data that count". I think it is fair to assume that most if not all of these trading systems suffer from the flaw that their excellent historical results are based on data that were used to design and tune the system. At this time, of course, my system is no different in this regard. The difference is that in the coming months I will be reporting in these pages "data that count" as they come in. The first "data that count" trades will be reported in next month's Status Report. In addition to collecting "data that count" under today's market conditions as they come in, I am also engaged in collecting historical "data that count" -- data prior to Nov. 16, 1990. It may take months to collect this data, as data must be keyed into my computer from each issue of Investor's Daily. I am hopeful that I will be able to acquire data at least back to 1987. As this historical data comes in, I will report the results in these pages. Will I re-design the DESI Trading System(tm) if it does not work well over time periods other than those used in its design? Perhaps, but if I do, I will have to start over in terms of collecting "data that count". The DESI Trading System(tm) in Three Steps In the next several sections, I present the new improved DESI Trading System(tm) in three steps: First, a discussion of how buy candidates are identified; second, a discussion of a new market timing system which is used to avoid buying when market conditions are unfavorable; and third, a discussion of the System's new sell strategy. Step 1: Identifying Buy Candidates This is the piece of the puzzle that has been in place for the longest time. Many months ago I invented a computerized selection process for identifying the hottest earnings surprises. Each day I key in data from Investor's Business Daily, and my computer program processes this data to produce a report of all DESI(tm) buy candidates. Most days, there are none. Occassionally there is one, and, rarely, there may be two or more on the same day. The program looks for the strongest market reactions to the strongest earnings reports; beyond that, the details are proprietary. Below is a complete table of all DESI(tm) buy candidates which were available to me when I revised the System in June and early July. The buy date is one day AFTER the earnings report is published in the newspaper -- this is typically two days after the company issues its press release. Many times, the stock will sell off a bit the day after a very strong up day, so this delay typically results in a lower buy price. This one-day delay is one of the recent improvements to the System. To give you some idea of how these buys performed, I have shown each stock's price 3 months later and the resulting gain or loss. STEP 1. DESI(tm) BUY CANDIDATES, NOV. 16, 1990 THROUGH JULY 22, 1992 Buy Buy 3 mos.Later Buy Buy 3 mos.Later Ticker Date Price Price G/L Ticker Date Price Price G/L BORL 1/18/91 39.00 57.25 +46.8% COLC 10/18/91 23.62 27.12 +14.8% IBM 1/22/91 119.62 107.88 -9.8% MRV 10/22/91 22.31 24.31 +9.0% TAVI 1/22/91 10.33 18.67 +80.6% RVAC 10/25/91 32.25 45.25 +40.3% ILCT 1/29/91 9.25 12.25 +32.4% EQIC 11/07/91 17.75 19.75 +11.3% DGII 1/31/91 15.25 19.25 +26.2% CSCO 11/11/91 26.75 40.62 +51.9% CSCO 2/12/91 14.81 15.25 +3.0% ECILF 11/11/91 23.25 28.75 +23.7% USHC 2/14/91 26.83 29.00 +8.1% FNIN 11/20/91 42.00 63.00 +50.0% ECILF 2/20/91 9.75 14.00 +43.6% RCDC 12/19/91 12.62 24.00 +90.1% TW 2/25/91 18.00 20.00 +11.1% SLR 12/24/91 17.25 21.75 +26.1% HON 2/26/91 57.50 56.75 -1.3% WHO 12/30/91 18.50 26.88 +45.3% UFCS 2/28/91 43.00 45.50 +5.8% CBEX 1/08/92 20.25 14.50 -28.4% FWC 3/01/91 32.75 27.75 -15.3% HELE 1/10/92 20.75 19.00 -8.4% LDDSA 3/07/91 20.50 18.00 -12.2% LINZ 1/14/92 33.75 33.50 -0.7% FULL 3/28/91 44.50 41.00 -7.9% MFC 1/20/92 22.12 21.00 -5.1% DREAF 4/10/91 16.50 10.00 -39.4% STCP 2/04/92 21.83 25.25 +15.7% CS 4/19/91 43.00 43.25 +0.6% PHSYA 2/06/92 27.00 28.12 +4.2% VAR 4/26/91 49.25 38.38 -22.1% DTM 2/14/92 11.50 15.62 +35.9% BPILF 5/30/91 38.75 26.75 -31.0% MRV 2/18/92 30.00 23.62 -21.2% SMLS 7/01/91 70.50 50.25 -28.7% OAR 2/25/92 47.50 59.50 +25.3% WAMU 7/19/91 27.75 33.50 +20.7% NWLIA 3/06/92 39.75 28.75 -27.7% USHC 8/05/91 31.50 31.25 -0.8% WHO 3/27/92 27.50 16.38 -40.5% STW 8/12/91 25.88 25.00 -3.4% MDB 4/09/92 13.75 12.88 -6.4% CAW 8/23/91 29.12 30.62 +5.2% FRCC 4/20/92 15.00 18.00 +20.0% TECD 8/23/91 23.00 25.50 +10.9% SUP 4/20/92 57.50 49.25 -14.3% BSBL 9/23/91 16.17 27.25 +68.6% XRIT 4/20/92 34.00 35.75 +5.1% SOME 10/09/91 15.25 19.25 +26.2% JSTN 4/21/92 20.67 27.50 +33.1% APS 10/15/91 41.00 39.88 -2.7% PMI 4/21/92 49.75 34.00 -31.7% MER 10/17/91 51.12 63.25 +23.7% IRWN 4/22/92 51.00 53.25 +4.4% MFC 10/17/91 21.50 21.50 0.0%* * Trades with a 0% gain are tallied as losses. Prices are split-adjusted where appropriate. Here are some statistics on these results, again assuming a holding period of 3 months: Step 1: Buy Candidates ------ Number of winners . . . 34 Number of losers. . . . 23 Percentage of winners . 60% Best trade. . . . . . . +90.1% Average winning trade . +27.0% Worst trade . . . . . . -40.5% Average losing trade. . -15.6% Average trade overall . +9.8% Step 2: Market Timing These statistics from Step 1 are very good, but the period of time covered was primarily a bull market. What about bear markets? The period of mid- February through mid-June of 1992 was a bear market for growth stocks, as the NASDAQ Composite fell 15% between Feb. 12 and June 26, with growth stocks falling even more than that. The DESI(tm) buy candidate stocks did not escape the pounding. 5 out of 9 DESI(tm) buy candidates in March and April turned out to be losers 3 months later, as can be seen from the data above. The market timing portion of the DESI Trading System(tm) attempts to avoid at least some of these losing trades by not buying when market conditions are unfavorable. A few months ago, in Vol.1 #50, I introduced a market timing system. The idea was to create an index of DESI(tm) buy candidates, and from the trend of that index decide whether it is safe to buy or not. This worked well, but I have since improved upon this idea with the DESI-3(tm) timing system. The DESI-3(tm) timing system constructs and tracks an index of only the three most recent DESI(tm) buy candidates. If this index is behaving well, the timing system gives a green light; otherwise, a red light is signalled. Each trading day will either be a green-light day or a red-light day. A red light does NOT mean "sell", only "do not buy". The exact details of how the green light/red light decision is made are proprietary. In order to buy a stock, the System insists that a green light be in effect for two days running: both on the buy date and on the previous trading day. My research indicates that this DESI-3(tm) timing system yields better results than the original DESI(tm) index. Applying this DESI-3(tm) timing system to the 57 buy candidates above results in throwing out 35 trades -- 17 losers and 18 winners -- due to a red light. This may not sound like a great idea, throwing out 18 winners and 17 losers, but it is! Look at what we are left with: 16 winners and 6 losers! Thus, the timing system has improved our winners-to-losers ratio from a hair under 3-to-2 (34-to-23) to a hair better than 5-to-2 (16-to-6). Stated another way, the timing system has improved the percentage of winners from 60% to 73%. I find it particularly significant and gratifying that, during the bearish months of March and April 1992, ALL NINE DESI(tm) buy candidates were rejected. The 22 trades which remain (after throwing out 35) are given below, with the resulting statistics. STEP 2. DESI(tm) BUY CANDIDATES WITH GREEN LIGHTS Buy Buy 3 mos.Later Buy Buy 3 mos.Later Ticker Date Price Price G/L Ticker Date Price Price G/L IBM 01/22/91 119.62 107.88 -9.8% CAW 08/23/91 29.12 30.62 +5.2% TAVI 01/22/91 10.33 18.67 +80.6% TECD 08/23/91 23.00 25.50 +10.9% ILCT 01/29/91 9.25 12.25 +32.4% MER 10/17/91 51.12 63.25 +23.7% DGII 01/31/91 15.25 19.25 +26.2% MFC 10/17/91 21.50 21.50 0.0% USHC 02/14/91 26.83 29.00 +8.1% COLC 10/18/91 23.62 27.12 +14.8% ECILF 02/20/91 9.75 14.00 +43.6% CSCO 11/11/91 26.75 40.62 +51.9% TW 02/25/91 18.00 20.00 +11.1% ECILF 11/11/91 23.25 28.75 +23.7% HON 02/26/91 57.50 56.75 -1.3% WHO 12/30/91 18.50 26.88 +45.3% FWC 03/01/91 32.75 27.75 -15.3% HELE 01/10/92 20.75 19.00 -8.4% FULL 03/28/91 44.50 41.00 -7.9% PHSYA 02/06/92 27.00 28.12 +4.2% WAMU 07/19/91 27.75 33.50 +20.7% DTM 02/14/92 11.50 15.62 +35.9% Prices are split-adjusted where appropriate. Step 1: Step 2: Buy DESI-3(tm) Candidates Timing ------ ------ Number of winners . . . 34 16 Number of losers. . . . 23 6 Percentage of winners . 60% 73% Best trade. . . . . . . +90.1% +80.6% Average winning trade . +27.0% +27.4% Worst trade . . . . . . -40.5% -15.3% Average losing trade. . -15.6% -7.1% Average trade overall . +9.8% +18.0% Step 3: When to Sell The statistics from Step 2 are very good indeed, but are based on the simplistic assumption that a stock is sold exactly 3 months after it is bought. Surely a more sophisticated sell strategy would improve the results even further. For the purpose of evaluating sell strategies, I used all DESI(tm) buy candidates, regardless of the green light/red light timing. This allowed me to run various sell strategies over many more trades, including many more losing trades. My previous sell strategy -- the one I presented in Vol. 1 #50 -- turns out to be worse than the simple-minded "hold exactly 3 months then sell" idea. This previous sell strategy misses out on many big gains by taking profits too soon. I tried various sell strategies, but, surprisingly, I found very few which improved upon the simple 3-month holding period. For example, adding a stop loss feature increases the number of losers and decreases the number of winners, resulting in a lower overall average gain. I tried other constant holding periods -- 2 months, 4 months, 6 months, etc. -- and found that 3 months was near ideal. I also tried a number of other ideas, but most failed to improve upon the simple 3-month approach. The sell strategy I finally chose is only a slight variation on "hold exactly 3 months". The basic idea is to hold the stock for 3 months, then wait for evidence of a short-term peak. The details of how a short-term peak is detected are proprietary, but the maximum holding period is three and a half months. So the sell always comes 3 to 3.5 months after the buy. This is the best strategy I have found so far. Here, then, are the results of applying this sell strategy to the stocks bought in Step 2. This is the complete DESI Trading System(tm). STEP 3. SELL STRATEGY -- THE COMPLETE DESI TRADING SYSTEM(tm) Buy Buy Sell Sell Gain/ Buy Buy Sell Sell Gain/ Ticker Date Price Date Price Loss Ticker Date Price Date Price Loss IBM 1/22/91 119.62 5/07 101.88 -14.8% CAW 8/23 29.12 12/06 29.75 +2.2% TAVI 1/22/91 10.33 4/26 21.50+108.1% TECD 8/23 23.00 11/29 27.25 +18.5% ILCT 1/29/91 9.25 5/07 13.75 +48.7% MER 10/17 51.12 1/27 60.62 +18.6% DGII 1/31/91 15.25 5/06 18.25 +19.7% MFC 10/17 21.50 1/30 23.88 +11.1% USHC 2/14/91 26.83 5/28 35.25 +31.4% COLC 10/18 23.62 1/22 25.50 +7.9% ECILF 2/20/91 9.75 5/29 14.62 +50.0% CSCO 11/11 26.75 2/18 41.12 +53.7% TW 2/25/91 18.00 6/03 20.38 +13.2% ECILF 11/11 23.25 2/14 29.75 +28.0% HON 2/26/91 57.50 6/05 62.50 +8.7% WHO 12/30 18.50 4/09 19.12 +3.4% FWC 3/01/91 32.75 6/05 30.50 -6.9% HELE 1/10 20.75 4/24 22.25 +7.2% FULL 3/28/91 44.50 7/05 44.50 0.0% PHSYA 2/06 27.00 5/22 27.06 +0.2% WAMU 7/19/91 27.75 10/31 29.50 +6.3% DTM 2/14 11.50 5/26 14.38 +25.0% Prices are split-adjusted where appropriate. Step 1: Step 2: Step 3: Buy DESI-3(tm) Sell Candidates Timing Strategy ------ ------ ------ Number of winners . . . 34 16 19 Number of losers. . . . 23 6 3 Percentage of winners . 60% 73% 86% Best trade. . . . . . . +90.1% +80.6% +108.1% Average winning trade . +27.0% +27.4% +24.3% Worst trade . . . . . . -40.5% -15.3% -14.8% Average losing trade. . -15.6% -7.1% -7.2% Average trade overall . +9.8% +18.0% +20.0% Notice in particular that this sell strategy improves the winners-to-losers ratio significantly, from 73% (16-to-6) to 86% (19-to-3). In other words, three losers turned into winners by virtue of waiting a little longer for a short-term rally. Also, TAVI, the +80.6% best winner of step 2 becomes a double, with a gain of 108.1%. How Do I Plan to Use the DESI Trading System(tm)? My intention is to follow the DESI Trading System(tm) with my own money, primarily in the Model Portfolio. I have already begun to do this. As of July 31, the DESI Trading System(tm) had issued two buys since the design changes I made in June and early July. I purchased both stocks on the very day dictated by the System. There may be times when I purchase DESI(tm) stocks in the Stock+Options Account, and there may be times when I override the System. For example, I might sell on a disappointing earnings report. In such a case, I would issue two sell alerts: one on the day I actually sell, and a second on the day the DESI Trading System(tm) issues a sell. In most cases, however, I expect to follow the System. I will continue to engage in other trading outside of the System; however, I plan to limit such other trading to the Stock+Options Account. Also, I will continue the new feature inaugurated in Vol.1 #77 of alerting you to other trading ideas -- ideas which do not fall into my usual categories. Performance Review: The Details Table 1 shows the status of all "data that count" DESI Trading System(tm) trades. There are no completed trades as of July 31, so this is not very significant. Next month's Status Report will include the first "data that count" trades. Tables 2 and 3 review the status of the Model Portfolio as of June 30 (Table 3 gives performance numbers). Tables 4 and 5 do the same for the Stock+Options Account. Past Status Reports have given two additional tables, one listing all completed stock trades so far this year, and the other listing all completed option trades. Due to the length of these tables, I am no longer including them. I will send this data by US mail upon request; please enclose a self- addressed stamped envelope. All tables reflect the state of trading as of July 31, 1992. Table 1: DESI Trading System(tm) Stocks -- "Data That Count"** DESI-buy* DESI-sell* Jul31 Gain/ Symbol Company Name Date High Date Low Close Loss OAR Ohio Art Co. 4/28 50.00 56.00 +12.0% *** LANTF Lannet Data Comm. Ltd. 5/08 28.25 29.50 +4.4% *** BJICA Ben & Jerry's Homemade 5/11 22.50 28.25 +25.6% *** PHSYA PacifiCare Health Sys. 7/31 35.50 *** UWSI United Wisconsin Svcs. 7/31 39.50 *** Statistics on Completed Trades: No completed trades as of July 31. * Stocks are assumed to be bought at the day's high and sold at the day's low. ** "Data That Count" means that these trades were not used in the design of the DESI Trading System(tm). *** These stocks are currently held by the System. Table 2. Model Portfolio Transaction Log Sym- B/ # Pro- Cash Date bol Company Name S Shrs Price Cost** ceeds** Balance From previous month . . . . . . . . . . . . . . . . . $16,361 7/15 CFFS Columbia First Bank B 350 14.75* $5,188 $11,172 7/20 WHO Waterhouse Inv Svcs S 150 16.00* $2,388 $13,560 7/29 LCSI LCS Industries S 500 7.25* $3,607 $17,167 7/31 PHSYA PacifiCare Health Sys B 140 35.00* $4,925 $12,243 7/31 UWSI United Wisconsin Svcs B 190 36.75* $7,017 $5,225 End-of-month cash balance. . . . . . . . . . . . . . . . . . . . $5,225 * In cases where I actually bought or sold a model portfolio stock, I have shown my transaction price and marked the price with an asterisk. In cases where I did not trade the stock, the day's closing price is shown. This applies to Tables 2 and 3. ** Includes transaction cost of 0.5% Table 3. Model Portfolio Status Sym- Purch # Purch Orig Jul31 Jul31 Gain/Loss bol Company Name Date ShrsPrice Cost** Price Value $$ % CS Cabletron Systems 4/2 165 54.00* $8,955 54.50 $8,993 $38 +0.4% IFMX Informix 5/11 200 34.00* $6,834 37.75 $7,550 $716 +10.5% CLCDF Clearly Canadian 5/12 500 18.00* $9,045 15.63 $7,813 ($1,233)-13.6% CFFS Columbia 1st Bank 7/15 350 14.75* $5,188 14.75 $5,163 ($26) -0.5% PHSYA PacifiCare Health 7/31 140 35.00* $4,925 35.50 $4,970 $46 +0.9% UWSI United Wisconsin 7/31 190 36.75* $7,017 38.50 $7,315 $298 +4.2% Total value of stocks held . . . . . . . $41,964 $41,803 ($161) -0.4% End-of-month cash balance. . . . . . . . . . . . . . . $5,225 One-month portfolio gain . . . . . . . . $45,264 $47,028 $1,763 +3.9% Year-to-date portfolio gain. . . . . . . $50,000 $47,028 ($2,972) -5.9% * See footnote, Table 2. ** Includes transaction cost of 0.5% Table 4. Stock+Option Account(*) Transaction Log Sym- B/ # Pro- Cash Date bol Company Name S Shrs Price Cost** ceeds** Balance Long Transactions and Covered Shorts Cash balance from last month (excl. cash from open shorts) . . . $2,963 7/6 CCR Countrywide Credit B 100 38.88 $3,916 ($952) 6/25 XOMA Xoma Corp.(short) S 800 13.00 $10,359 $9,407 7/17 XOMA Xoma Corp.(cover) B 800 14.00 $11,241 ($1,834) 7/17 ARW Arrow Electronics B 300 21.50 $6,484 ($8,318) 7/20 WHO Waterhouse Inv Svcs S 400 16.00 $6,363 ($1,955) 7/29 FTXTD FreeptMcMoran Aug20put S 10 1.00 $965 ($990) 7/29 RUS Russ Berrie & Co. B 150 28.50 $4,305 ($5,295) 7/29 C Chrysler Corp. B 200 22.00 $4,431 ($9,726) End-of-month cash balance (excl. cash from open shorts). . . . . ($9,726) Short sells: 7/17 USS US Surgical S 100 98.00 $9,772 $9,772 * This is an actual, personal account of the editor. Table 5. Stock+Options Account(*) Status Sym- Purch # Purch Orig Jul31 Jul31 Gain/Loss bol Company Name Date Shrs Price Cost Price Value $$ % Long Positions: CS Cabletron Systems 4/2 55 54.00 $2,997 54.50 $2,998 $1 +0.0% CS Cabletron Systems 4/2 55 51.00 $2,807 54.50 $2,998 $191 +6.8% MDB Prof. Bancorp 4/14 200 15.00 $3,031 11.75 $2,350 ($681) -22.5% CSFCB CSF Holdings Inc. 6/9 250 22.00 $5,529 34.00 $8,500 $2,971 +53.7% MDB Prof. Bancorp 6/11 200 12.50 $2,531 11.75 $2,350 ($181) -7.2% CCR Countrywide Credit 7/6 150 25.92 $3,916 29.63 $4,444 $528 +13.5% ARW Arrow Electronics 7/17 300 21.50 $6,484 20.25 $6,075 ($409) -6.3% RUS Russ Berrie & Co. 7/29 150 28.50 $4,305 28.50 $4,275 ($30) -0.7% C Chrysler Corp. 7/29 200 22.00 $4,431 21.88 $4,375 ($56) -1.3% Total value of long positions held . . . . . . . . . . . $38,364 End-of-month cash balance, excl. cash from open shorts . ($9,726) Portfolio value, long portion. . . . . . . . . . . . . . $28,638 Short Positions: USS US Surgical 7/17 -100 98.00 ($9,772) 91.00 ($9,100) $672 +6.9% Summary of short positions . . . . . . . ($9,772) ($9,100) $672 +6.9% Account Summary: Total portfolio value (long portion+net gain on shorts) $29,310 One-month portfolio gain . . . . . . . . $27,400 $29,310 $1,910 +7.0% Portfolio gain since inception (3/31/92) $33,001 $29,310($3,691) -11.2% * This is an actual, personal account of the editor. -------------- Footnotes (1) On the day that the System says "buy", that day's high is used as the buy price. Likewise, the day's low is used as the sell price when a "sell" is given. It is a bit of a handicap, since typically one can get a better price than that. On the other hand, the ignoring of commissions artificially boosts the performance. This handicap and this boost roughly cancel each other out. (2) The program I am writing to track a $25,000 investment using the DESI Trading System(tm) will take commissions into account, using the commission schedule of a deep discount broker. It will also have built into it the same handicap mentioned in footnote (1), in order to account for slippage and the bid/ask spread. Since this program has this handicap without the artificial boost of ignoring commissions, the results reported will be a bit pessimistic. ---------------------------------------------------------------------------- The Aggressive Trader(tm) is edited and published at irregular intervals, but at least monthly, by Kenneth L. Deen ("Ken Deen"), 1339 Virginia Road, Montecito, California 93108, (805) 505-2039, and is distributed free of charge to all interested parties. Ken Deen reserves the right to begin charging a subscription fee at any time. Ken Deen, is employees, affiliates, and/or clients may have positions in securities recommended herein and may make additional purchases and/or sales in these securities. Ken Deen frequently reveals in this publication the holdings, trading activity, and performance of certain of his personal brokerage accounts. However, he reserves the right to discontinue this practice at any time without notice. Recommendations made in this publication involve risk and may result in losses. Readers should not assume that recommendations will be profitable or will equal past performance. Neither The Aggressive Trader(tm) nor Ken Deen shall be liable in any manner for losses of any kind. The information in this publication is collected from sources believed to be reliable, but neither the accuracy nor the completeness of this information can be guaranteed in any way. Reporduction in whole or in part without the express written consent of Ken Deen is strictly prohibited. Copyright 1992 Kenneth L. Deen. -END-