[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Omega Research



PureBytes Links

Trading Reference Links

Gordon, thanks for the reply and info - Richard

At 04:17 AM 7/26/97 GMT, Guy Gordon wrote:
>I firmly believe that the period of a Moving Average should be
>determined by the stock data itself.  You want it long enough to
>smooth out unwanted variation, but not so long that it erases the
>variation you are interested in.
>
>I watch 26-day and 50-day Moving Averages to verify my trendlines.
>I use RSI(9) and STOCH(9,4,4) to make my trades.  If the STOCH(9) is
>too slow, I use STOCH(5,3,3).
>
>But I can't prove (even to my own satisfaction) that these are the
>best numbers to use.  Theoretically, you should be able to use an FFT
>(Fast Fourier Transform) to find the best period.  But I have not had
>any luck using that.
>
>On Fri, 25 Jul 1997 20:02:37 -0700, R Rodgers wrote:
>
>>Hi Gordon, I'm curious concerning the amount of days you use for the RSI
>>and Stochastics indicators.  I also use these indicators to confirm each
>>other and use them only for certain jobs (for dips only or rallies only and
>>depending on the trend - will use one oscillator for dips and others for
>>rallies, and certain ones for confirmation, etc.)  For longer term, I've
>>used 25-day Momentum, 25-day RSI, and 25-day Percent R. Then divide the
>>Type 1, Type 2 divergences, and rallies and dips between them.  Thanks for
>>your time - Richard
>>
>>At 10:00 PM 7/25/97 GMT, you wrote:
>>>The thing about those charts is how clear the buy is.
>>>
>>>The Stochastic gives some very good buy and sell signals.  Perhaps too
>>>many of them.  The RSI gives fewer signals.  I try to use it to verify
>>>the Stoch signals.
>>>
>>>I have noticed this about RSI:  When the stock is in an uptrend
>>>channel, the RSI gives good SELL signals, but will not give you a BUY
>>>signal to get back into the stock.  I use the Stochastic for the buy.
>>>
>>>Similarly, when the stock is in a downtrend, the RSI will give you a
>>>good BUY signal, but not a sell.
>>>
>>>Perhaps the problem is that the horizontal lines at 80 and 20 are not
>>>flexible enough.  Maybe they should be in different places based on
>>>the stock trend.
>>>
>>>On Fri, 25 Jul 1997 14:03:49 -0400, you wrote:
>>>
>>>>     I looked at your charts.  Thanks!
>>>>
>>>>Jim
>>>>
>>>>----------
>>>>From: Guy Gordon <gordon@xxxxxxxxxxx>
>>>>To: metastock-list@xxxxxxxxxxxxx
>>>>Subject: Re: Breakout Alerts: CUBE and PAIR
>>>>Date: Thursday, July 24, 1997 9:03 PM
>>>>
>>>>If anyone wants to see my charts on PAIR, CUBE, QNTM, and AGPH, they
>>>>are posted at:
>>>>
>>>>http://www.white-crane.com/pair.htm
>>>>http://www.white-crane.com/cube.htm
>>>>http://www.white-crane.com/qntm.htm
>>>>http://www.white-crane.com/agph.htm
>>>>
>>>>On Tue, 22 Jul 1997 21:00:03 GMT, I wrote:
>>>>
>>>>>Just a quick note:  take a look at C-Cube Microsystems (CUBE) and
>>>>>Pairgain Tech (PAIR).  Both of these look like good buys.
>
>