|
PureBytes Links
Trading Reference Links
|
I completely agree! How much predictive behavior can you squeeze out of two variables (time and price)? Add volume and you've got three, and that's still a small number. And yet, folks bang their head to create complex indicators with multi-colored squiggles and flashing doohickies, trying to find the next price move.
Here's an exercise: try to predict next week's weather with three variables.
And some advice: put away the complex technical analysis trading software for a while, come back to it later. In an uptrend, by good stocks using a backtested screen (that's where there's plenty of work to be done, rather than in the technical analysis arena). Buy on pullbacks to the 10MA or 20MA. Exit when they hit your trailing stop or exit target. Can it be any simpler?
Andy
--- In equismetastock@xxxxxxxxxxxxxxx, superfragalist <no_reply@xxx> wrote:
>
> Basically there are only two variables for a specific stock that are available to work with when using TA. There is price and volume history.
>
> I'm always surprised at how much information TA users expect that those two variables are going to provide. No matter what kind of rocket science, engineering mathematics, lunar equations or Elliot waves are applied to a stock's price and volume information, nothing more can be rung out of that than can be found using simple arithmetic or elementary statistics.
>
> In addition, the future price of a stock is only partially influenced by its recent historical price and volume. The economy, current events, the overall market and a variety of other factors influence the price of a stock. As a simple example, let's suppose a stock was in a straight line upward progression that looked like it fit the linear regression line of the price data perfectly. The following day a nuke is exploded in Seattle and Starbucks headquarters is wiped out. The price is not going to stay on the linear regression path upward while all the other stocks in the market are tanking.
>
> Tomorrow's price of a stock can only be defined in terms of a probability. That probability is made up both the previous price and volume history and the probability of external events that influence not only that specific stock but the market as a whole. If only the price and history information is used to define the range of possible prices tomorrow, or months from now, then the end points of the price range have to be very broad.
>
> The future price of a stock should be defined using a confidence interval. Which means the price tomorrow, or at any point in the future, is going to fall between X and Y 90% (or 95% or 99% etc) of the time. The problem is if the current price is $10, then the end points on a 90% confidence interval would likely be $8 to $12, and even then 10% of the time the projected stock price is going to fall outside of the $8 to $12.
>
> Since the confidence interval is based only on the historical price and volume of the stock, and does not include the influence of any external events, that confidence interval is imprecise and may not even be statistically relevant, depending on the influence of external events at that moment in time.
>
> What that means is the X and Y end points of the confidence interval are so far apart, the information is irrelevant for making decisions.
>
> The reason some people day trade is they want to remove the longer term impact of external events on the likely price of a stock. They are betting that some time during the day based only on previous price and volume information, the price of the stock will move away from the entry price and closer to X if they are long or closer to Y if they are short.
>
> Even if all the influence of external events were completely removed from a stock's price movement, statistically the price has an equal chance of moving closer to X or to Y from the mean. If the confidence interval were precisely calculated and the entry price executed exactly at the mean within the confidence interval, then the chances the stock would go up or down are 50%, which is equal to tossing a coin.
>
> TA adds very little information about the future value of a stock because TA has only price and volume information to look at.
>
> I've written articles on the relevance of TA and why what someone does before they apply TA to a stock has more influence on the probability of a good outcome after the trade than using TA itself. I've also explained why mechanical trading systems, in general, have limited life expectancy with performance that is going to degrade over time. When the history of any mechanical system is dissected, its performance is going to be inconsistent. Only under specific market conditions is it going to perform well. The rest of the time it will be erratic.
>
> Some people like complexity and believe that the more complex a system is (Elliot waves and Gann Charts), the better it must be. Others believe that applying acoustical mathematics to price and volume information cleans up the noise and makes trading easier, and some people believe that the phases of the moon exert a force on the minds of fund mangers, which then influences the price of stocks.
>
> Okay, whatever works for you. For me, the money is made with arithmetic before and during the trade. Simple stuff.
>
> Super
>
>
>
>
>
>
>
>
> --- In equismetastock@xxxxxxxxxxxxxxx, "Ed Hoopes" <reefbreak_sd@> wrote:
> >
> > I have backtested several methods of projecting the current trend to the next day. I have fitted the last n bars to a linear least squares fit, a quadratic least squares fit, and a fourier transform of the last n days. Each of these results in an equation that can be solved for one bar in the future.
> >
> > My backtests show that none are profitable.
> >
> > My conclusion is that there is a tendency for humans to look at a chart and think that the price trend contains some inertial mass, similar to a car moving down the road. Unfortunately, the price trend of a stock has zero inertial mass.
> >
> >
> >
> > --- In equismetastock@xxxxxxxxxxxxxxx, pumrysh <no_reply@> wrote:
> > >
> > > Krishna,
> > >
> > > No one can give you tomorrow's value. To do so is projection or speculation and certainly not valid.
> > >
> > > There was a users group some time back which developed a formula for predicting the closing price needed for the signal line to cross the MACD.
> > >
> > > Not very helpful when you think about it since the close does not come until the very end of the trading day.
> > >
> > > Preston
> > >
> > >
> > >
> > > --- In equismetastock@xxxxxxxxxxxxxxx, "k_krishna79" <k_krishna79@> wrote:
> > > >
> > > > Normally the moving average formula for metastock is mention below:
> > > >
> > > > Mov(Data Array, Periods, E S T TRI VAR W VOL)
> > > >
> > > >
> > > > Normally the moving average that is calculated is for the next date, can i have the line ahead, what i mean is the leading moving average.
> > > >
> > > > Kindly help.
> > > >
> > > > Regards,
> > > > KRISHNA
> > > >
> > >
> >
>
------------------------------------
Yahoo! Groups Links
<*> To visit your group on the web, go to:
http://groups.yahoo.com/group/equismetastock/
<*> Your email settings:
Individual Email | Traditional
<*> To change settings online go to:
http://groups.yahoo.com/group/equismetastock/join
(Yahoo! ID required)
<*> To change settings via email:
mailto:equismetastock-digest@xxxxxxxxxxxxxxx
mailto:equismetastock-fullfeatured@xxxxxxxxxxxxxxx
<*> To unsubscribe from this group, send an email to:
equismetastock-unsubscribe@xxxxxxxxxxxxxxx
<*> Your use of Yahoo! Groups is subject to:
http://docs.yahoo.com/info/terms/
|