To: Steve 667 who wrote (2088 ) 12/12/1997 9:48:00 AM From: Loren Read Replies (4) | Respond to of 60323
Steven - << Since you are oooh sooo sensitive I will try to ever so politely put it another way: >> Ok, Steven, you are very polite... now if you'll just work on not patronizing me, we'll get along just great...:-) Now, I notice you have e-mail... what kinds of software do you use? The reason I ask is, I would be happy to send you a picture of a chart to show you what I'm talking about... it would be worth at least 1000 of my words. In the meantime, I'll try to portray what I'm talking about in words... a. I use DMI and MACD indicators... when I chart them, I use weekly price data in one window, and daily price data in another window. I use daily charts for reference only... I use weekly charts for trading. The weekly timeframe just fits my trading style better (since I have a day job already). b. Both DMI and MACD are trend indicators, as opposed to oscillators (like RSI or stochastic). Both attempt to predict when trends have been established. They don't try to call something overbought or oversold (expensive or cheap). c. MACD uses differences of moving averages to highlight/accentuate changes in price movement. To me, it's very analogous to determining acceleration and velocity by sampling the position of a moving object (like a car). Once you know which way a car is accelerating or decelerating, you can make some pretty good predictions about where the car is going in the near-term future. d. DMI simply looks at the frequency and strength of upward vs. downward price movements, and attempts to show whether the stock is trending up, down, or not at all. e. How long does it take for a trend to be 'established'? It depends on whether you are charting daily data or weekly data, and it depends on the relative volatility of the stock. All of these indicators are affected by noise (like Asian markets crashing around you). The better to signal to noise ratio, the sooner you can pick out the signal! Let me just illustrate how SNDK has worked in the last two years, using my DMI indicators... If you had traded SNDK based on weekly data, you would made 5 trades during the two years, and made money on only two of them (40% right) 1. September 96 - lost ~1 $/shr during a 5 week trade. 2. November 96 - lost ~1 $/shr during a 6 week trade. 3. January 97 - lost ~0.50 $/shr during an 8 week trade. 4. April 97 - made ~0.50 $/shr during a 4 week trade. 5. May-Oct 97 - made ~$18 $/shr during a 23 week trade. This gives a net gain of ~$16/shr. With a starting point of 14 1/4 (the first buy signal in 9/96), and considering that I only had my money tied up in SNDK for 46 weeks out of the two years, that's a rate of return of well over 100%. Even being right only 40% of the time on the stock! Now, if I had bought in in September 96 based on that first buy signal (at 14 1/4) and just held until now, I'd be up 5 1/2 points. Over the period from 9/96 to now, that's about 50%. Not bad either, but not as good. The key to this type of trading is to not leave your money in something that's going down or sideways. When it's going down, you're losing capital. When it's going sideways, you're losing the use of your money in another trade. Granted, I didn't actually do this with SNDK. This is just an example. Don't flame me for that, I didn't know about SNDK until just a few weeks ago. I promise to you, though, I have done the same exact thing with Presstek (PRST) and Zoltek (ZOLT) and Cisco (CSCO) and others... Now, I reiterate the offer to show you pictures of the charts at some point. That way you can see the indicators show the buy and sell signals. And I'll take up the offer to let this thread know when I see the next buy signal, assuming I am still watching this stock. And I probably will watch it for some time, given its excellent long-term prospects. Loren