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

Re: Time Series Forecast Formula (part1)



PureBytes Links

Trading Reference Links

<x-html><!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
<HTML><HEAD>
<META content="text/html; charset=iso-8859-1" http-equiv=Content-Type>
<META content="MSHTML 5.00.2614.3401" name=GENERATOR>
<STYLE></STYLE>
</HEAD>
<BODY>
<DIV><FONT size=2>
<DIV><FONT size=2>The EPMA(14)-indicator you have mentioned is the same as the 
Linnear Regresion Trendline(14)-indicator</FONT></DIV>
<DIV><FONT size=2>build-in MetaStock. </FONT><FONT size=2>But it is not the same 
as the </FONT><FONT size=2>TSF(14)-indicator or the TSF(14)-function. 
</FONT></FONT><FONT size=2>Adam is also</FONT></DIV>
<DIV><FONT size=2>right about the TSF-function's existance in MetaStock, as well 
as that there is also a </FONT><FONT size=2>TSF-indicator 
build-in.</FONT></DIV></DIV>
<DIV><FONT size=2></FONT>&nbsp;</DIV>
<DIV><FONT size=2>The Custom-2 indicator printed below </FONT><FONT size=2>will 
give you </FONT><FONT size=2>the underlying calculation method for both 
the</FONT></DIV>
<DIV><FONT size=2>TSF-indicator and the TSF-function, </FONT><FONT size=2>eg 
the&nbsp;Time Series Forecast-indicator as well as that it is the 
same</FONT></DIV>
<DIV><FONT size=2>calculation&nbsp;underlay </FONT><FONT 
size=2>for&nbsp;</FONT><FONT size=2>the&nbsp;Time Series Forecast-function, 
eg&nbsp;you can also use this (Custom-2) formula for</FONT></DIV>
<DIV><FONT size=2>creating any other </FONT><FONT size=2>formulas.</FONT></DIV>
<DIV><FONT size=2></FONT>&nbsp;</DIV>
<DIV><FONT size=2>Since you are keen to include the TRIX, have&nbsp;printed 
RIGHT AT THE BOTTOM of this mail, a few previous</FONT></DIV>
<DIV><FONT size=2>TRIX-mails on this S&amp;C-article. </FONT><FONT size=2>From 
the in their mail used Filter, I have created an indicator for 
MetaStock</FONT></DIV>
<DIV><FONT size=2>v6.5, eg the "</FONT><FONT size=2><FONT size=2>TIMESERIES TRIX 
- Joe Luisi", </FONT>which is also printed </FONT><FONT size=2>(now&nbsp;right) 
below.</FONT></DIV>
<DIV><FONT size=2>Both the 2-mails(right at the bottom) </FONT><FONT size=2>on 
the Joe Luisi's S&amp;C's article and </FONT><FONT size=2>the indicator(right 
below) stemming</FONT></DIV>
<DIV><FONT size=2>from these mails, requires </FONT><FONT size=2>you to use 
</FONT><FONT size=2>weekly data to get the&nbsp;best (Trend) results (see GIF in 
mail-part2).</FONT></DIV>
<DIV><FONT size=2>Tho, find the daily version quit usable as well.</FONT></DIV>
<DIV><FONT size=2></FONT>&nbsp;</DIV>
<DIV><FONT size=2>I hope someday someone can explain to me the "</FONT><FONT 
size=2>forecasting" value of the TSF (highly presented in 
MetaStock),</FONT></DIV>
<DIV><FONT size=2>eg as in how I should use it in a parculiar way so as to 
finaly get some forcasting results, as for up till now, this</FONT></DIV>
<DIV><FONT size=2>all still </FONT><FONT size=2>remains&nbsp;a big mistery 
to&nbsp;me.</FONT></DIV>
<DIV><FONT size=2>I find Joe's and my own Trend-detection indicator (TM-Rally 
Meter Osc)</FONT><FONT size=2>&nbsp;giving the right (very reliable) 
information</FONT></DIV>
<DIV><FONT size=2>in&nbsp;Forecasting(see GIF in mail-part2). </FONT></DIV>
<DIV>&nbsp;</DIV>
<DIV><FONT size=2>Regards,<BR>Ton Maas<BR><A 
href="mailto:ms-irb@xxxxxxxxxxxxx";>ms-irb@xxxxxxxxxxxxx</A><BR>Dismiss the 
".nospam" bit (including the dot) when replying.</FONT></DIV>
<DIV>&nbsp;</DIV>
<DIV><FONT size=2>=======================================</FONT></DIV>
<DIV><FONT size=2>TIMESERIES TRIX - Joe Luisi</FONT></DIV>
<DIV>&nbsp;</DIV>
<DIV><FONT size=2>{published in S&amp;C - TASC article "Playing<BR>Trix" by Joe 
Luisi (June 1997) and to be used on weekly 
data}<BR>CLA:=TRIX(3);<BR>CLB:=Ref(TRIX(3),-1);<BR>CLC:=Mov(TRIX(3),8,TIMESERIES);<BR>CLD:=Ref(Mov(TRIX(3),8,TIMESERIES),-1);<BR>SHORT:=When(CLA,&gt;,CLC) 
AND When(CLB,&lt;,CLD) AND<BR>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; When(CLA,&lt;,0)AND 
When(CLA,&gt;,-2);<BR>LONG:=When(CLA,&lt;,CLC) AND When(CLB,&gt;,CLD) 
AND<BR>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; When(CLA,&gt;,0)AND 
When(CLA,&lt;,+2);<BR>If(LONG&gt;0,+1,<BR>If(SHORT&gt;0,-1,PREVIOUS))</FONT></DIV>
<DIV><FONT size=2><FONT size=2></FONT></FONT>&nbsp;</DIV>
<DIV><FONT size=2><FONT 
size=2>=======================================</FONT></FONT></DIV>
<DIV><STRONG>Time Series Forecast</STRONG></DIV>
<DIV>&nbsp;</DIV>
<DIV><FONT size=2>
<DIV><FONT size=2><STRONG>Description</STRONG></FONT></DIV>
<DIV><FONT size=2>The Time Series Forecast indicator is based on the trend of a 
security's price over a specified time period.</FONT></DIV>
<DIV><FONT size=2>The trend is determined by calculating a linear regression 
trendline using the "least squares fit" method.</FONT></DIV>
<DIV><FONT size=2>The least squares fit technique fits a trendline to the data 
in the chart by minimizing the distance between</FONT></DIV>
<DIV><FONT size=2>the data points and the linear regression trendline.&nbsp; Any 
point along the Time Series Forecast is equal to</FONT></DIV>
<DIV><FONT size=2>the ending value of a Linear Regression trendline plus its 
slope.</FONT></DIV>
<DIV>&nbsp;</DIV>
<DIV><FONT size=2><FONT size=2><STRONG>Calculation</STRONG></FONT></FONT></DIV>
<DIV><FONT size=2>For example, the ending value of a Linear Regression trendline 
(plus its slope) that covers 10 days will</FONT></DIV>
<DIV><FONT size=2>have the same value as a 10-day Time Series Forecast.&nbsp; 
This differs slightly from the Linear Regression</FONT></DIV>
<DIV><FONT size=2>indicator (see Linear Regression Indicator) in that the Linear 
Regression indicator does not add the slope</FONT></DIV>
<DIV><FONT size=2>to the ending value of the regression line.&nbsp; This makes 
the TSF a bit more responsive to short term price</FONT></DIV>
<DIV><FONT size=2>changes.&nbsp;If you plot the TSF and the Linear Regression 
indicator side-by-side, you&#8217;ll notice that the TSF</FONT></DIV>
<DIV><FONT size=2>hugs the prices more closely than the Linear Regression 
indicator.</FONT></DIV>
<DIV>&nbsp;</DIV>
<DIV><FONT size=2>Rather than plotting a straight Linear Regression trendline, 
the Time Series Forecast indicator plots the</FONT></DIV>
<DIV><FONT size=2>ending values of multiple Linear Regression trendlines.&nbsp; 
The resulting Time Series Forecast indicator is</FONT></DIV>
<DIV><FONT size=2>sometimes referred to as a "moving linear regression" study or 
a "regression oscillator".</FONT></DIV>
<DIV><FONT size=2><BR></FONT></FONT><FONT size=2><FONT size=2><FONT size=2><FONT 
size=2><STRONG>Interpretation</STRONG></FONT></FONT></DIV></DIV>
<DIV><FONT size=2>The interpretation of a Time Series Forecast is similar to a 
moving average.&nbsp; However, the</FONT></DIV>
<DIV><FONT size=2>Time Series Forecast-indicator has two advantages over moving 
averages.</FONT></DIV>
<BLOCKQUOTE style="MARGIN-RIGHT: 0px">
  <DIV>
  <DIV style="MARGIN-RIGHT: 0px"><FONT size=2>1. Unlike a moving average, a Time 
  Series Forecast does not exhibit as much "delay".</FONT></DIV></DIV>
  <DIV style="MARGIN-RIGHT: 0px">&nbsp;&nbsp;&nbsp;&nbsp;<FONT size=2>Since the 
  indicator is "fitting" a line to the data points rather than averaging 
  them,</FONT></DIV>
  <DIV style="MARGIN-RIGHT: 0px">&nbsp;&nbsp;&nbsp;&nbsp;<FONT size=2>the Time 
  Series line is more responsive to price changes.<BR>2. As the name suggests, 
  the indicator can be used to forecast the next period's price.</FONT></DIV>
  <DIV style="MARGIN-RIGHT: 0px">&nbsp;&nbsp;&nbsp;&nbsp;<FONT size=2>This 
  estimate is based on the trend of the security's prices over the period 
  specified</FONT></DIV>
  <DIV style="MARGIN-RIGHT: 0px">&nbsp;&nbsp;&nbsp;&nbsp;<FONT size=2>(e.g., 20 
  periods).&nbsp; If the trend continues, the last point of the trendline (the 
  value of the</FONT></DIV>
  <DIV style="MARGIN-RIGHT: 0px">&nbsp;&nbsp;&nbsp;&nbsp;<FONT size=2>Time 
  Series Forecast) is forecasting the next period's 
  price.</FONT></FONT></FONT></DIV></BLOCKQUOTE>
<DIV><FONT size=2><STRONG>Parameters</STRONG></FONT></DIV>
<DIV><FONT size=2>The parameters for the Time Series Forecast are shown 
below.&nbsp;These parameters are specified at the</FONT></DIV>
<DIV><FONT size=2>time the indicator is plotted. You can edit the parameters of 
an existing plot by right-clicking on the</FONT></DIV>
<DIV><FONT size=2>indicator and choosing Properties from the shortcut 
menu.<BR></FONT></DIV>
<DIV><FONT size=2>-Time Periods.&nbsp; Enter the number of time periods to use 
when calculating the Time Series Forecast.</FONT></DIV>
<DIV><FONT 
size=2>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; 
&nbsp; The term "time periods" refers to days if the chart contains daily data, 
weeks for</FONT></DIV>
<DIV><FONT 
size=2>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; 
&nbsp;weekly data, etc.<BR>-Price Field.&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; Choose 
the price field (i.e., open, high, low, or close) to use when 
calculating</FONT></DIV>
<DIV><FONT 
size=2>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; 
&nbsp;the Time Series Forecast.</FONT></DIV>
<DIV><FONT size=2><STRONG>Custom-1</STRONG></FONT></DIV>
<DIV><FONT size=2>SYNTAX&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; tsf( DATA ARRAY, PERIODS 
)<BR>FUNCTION&nbsp; Calculates the predefined PERIODS Time Series Forecast 
indicator of DATA ARRAY. <BR>EXAMPLE&nbsp;&nbsp; The formula "tsf( CLOSE, 10 )" 
returns a 10-period Time Series Forecast of the closing prices.<BR>
<DIV><FONT size=2><STRONG></STRONG></FONT>&nbsp;</DIV>
<DIV><FONT size=2><STRONG>Custom-2</STRONG></FONT></DIV></FONT></DIV>
<DIV><FONT size=2>SYNTAX&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; 
LinearReg(WC(),14)+LinRegSlope(WC(),14)</FONT></DIV>
<DIV><FONT size=2></FONT>&nbsp;</DIV>
<DIV><FONT size=2>SEE ALSO&nbsp; Linear Regression Indicator and Forecast 
Oscillator.</FONT></DIV>
<DIV><FONT 
size=2>---------------------------------------------------------------------------------</FONT></DIV>
<DIV><STRONG>Linear Regression Indicator</STRONG></DIV>
<DIV>&nbsp;</DIV>
<DIV><FONT size=2>The Linear Regression indicator is based on the trend of a 
security's price over a specified time period.</FONT></DIV>
<DIV><FONT size=2>The trend is determined by calculating a linear regression 
trendline using the "least squares fit" method.</FONT></DIV>
<DIV><FONT size=2>The least squares fit technique fits a trendline to the data 
in the chart by minimizing the distance</FONT></DIV>
<DIV><FONT size=2>between the data points and the linear regression 
trendline.&nbsp;Any point along the Linear Regression</FONT></DIV>
<DIV><FONT size=2>indicator is equal to the ending value of a Linear Regression 
trendline.</FONT></DIV>
<DIV><FONT size=2>For example, the ending value of a Linear Regression trendline 
that covers 10 days will have the same</FONT></DIV>
<DIV><FONT size=2>value as a 10-day Linear Regression indicator.&nbsp; This 
differs slightly from the Time Series Forecast indicator</FONT></DIV>
<DIV><FONT size=2>(see Time Series Forecast) in that the TSF adds the slope to 
the ending value of the regression line.</FONT></DIV>
<DIV><FONT size=2>This makes the TSF a bit more responsive to short term price 
changes.</FONT></DIV>
<DIV><FONT size=2>If you plot the TSF and the Linear Regression indicator 
side-by-side, you&#8217;ll notice that the TSF hugs</FONT></DIV>
<DIV><FONT size=2>the prices more closely than the Linear Regression 
indicator.</FONT></DIV>
<DIV>&nbsp;</DIV>
<DIV><FONT size=2>Rather than plotting a straight Linear Regression trendline, 
the Linear Regression indicator plots the</FONT></DIV>
<DIV><FONT size=2>ending values of multiple Linear Regression 
trendlines.</FONT></DIV>
<DIV>&nbsp;</DIV>
<DIV><FONT size=2>Interpretation</FONT></DIV>
<DIV><FONT size=2>The interpretation of a Linear Regression indicator is similar 
to a moving average.&nbsp; However,</FONT></DIV>
<DIV><FONT size=2>the Linear Regression indicator has two advantages over moving 
averages.<BR>Unlike a moving average, a Linear Regression indicator does not 
exhibit as much "delay".</FONT></DIV>
<DIV><FONT size=2>Since the indicator is "fitting" a line to the data points 
rather than averaging them,</FONT></DIV>
<DIV><FONT size=2>the Linear Regression line is more responsive to price 
changes.<BR>The indicator is actually a forecast of the next periods 
(tomorrow&#8217;s) price plotted today.</FONT></DIV>
<DIV><FONT size=2>The Forecast Oscillator plots the percentage difference 
between the forecast price and the actual price.</FONT></DIV>
<DIV><FONT size=2>Tushar Chande suggests that when prices are persistently above 
or below the forecast price, prices</FONT><FONT size=2>&nbsp;can</FONT></DIV>
<DIV><FONT size=2>be expected to snap back to more realistic levels.&nbsp; In 
other words the Linear Regression indicator shows</FONT></DIV>
<DIV><FONT size=2>where prices should be trading on a statistical 
basis.</FONT></DIV>
<DIV><FONT size=2>Any excessive deviation from the regression line should be 
short-lived.</FONT></DIV>
<DIV>&nbsp;</DIV>
<DIV><FONT size=2>See also <FONT size=2>Linear Regression Trendline,&nbsp;<FONT 
size=2>Linear Regression&nbsp;</FONT>Slope and </FONT>Time Series Forecast. 
</FONT></DIV>
<DIV><FONT size=2><FONT 
size=2>---------------------------------------------------------------------------------</FONT></FONT></DIV>
<DIV><STRONG>Linear Regression Trendline</STRONG></DIV>
<DIV>&nbsp;</DIV>
<DIV><FONT size=2>Linear regression is a statistical tool used to predict future 
values from past values.</FONT></DIV>
<DIV><FONT size=2>In the case of security prices, it is commonly used as a 
quantitative way to determine the underlying</FONT></DIV>
<DIV><FONT size=2>trend and when prices are overextended.<BR>A Linear Regression 
trendline uses the least squares method to plot a straight line through 
prices</FONT></DIV>
<DIV><FONT size=2>so as to minimize the distances between the prices and the 
resulting trendline.<BR></FONT></DIV>
<DIV><FONT size=2>See Linear Regression Trendline and Raff Regression Channels 
for more information on</FONT></DIV>
<DIV><FONT size=2>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; &nbsp;the Linear Regression 
parameters.</FONT></DIV>
<DIV><FONT size=2><FONT size=2><FONT 
size=2>---------------------------------------------------------------------------------</FONT></FONT></FONT></DIV>
<DIV><STRONG>Linear Regression Slope</STRONG></DIV>
<DIV>&nbsp;</DIV>
<DIV><FONT size=2>The Linear Regression method provides several useful outputs 
for technical analysts, including the Slope.</FONT></DIV>
<DIV><FONT size=2>The Slope shows how much prices are expected to change per 
unit of time.&nbsp;Some may remember this</FONT></DIV>
<DIV><FONT size=2>as &#8220;rise over run&#8221;.</FONT></DIV>
<DIV><FONT size=2><BR>Interpretation</FONT></DIV>
<DIV>&nbsp;</DIV>
<DIV><FONT size=2>It is helpful to consider Slope in relation to r-squared (see 
r-squared).&nbsp; While Slope gives you the general</FONT></DIV>
<DIV><FONT size=2>direction of the trend (positive or negative), r-squared gives 
you the strength of the trend.</FONT></DIV>
<DIV><FONT size=2>A high r-squared value can be associated with a high positive 
or negative Slope. <BR>When the Slope of the trend first becomes significantly 
positive, you could open a long position.</FONT></DIV>
<DIV><FONT size=2>You could sell, or open a short position when the Slope first 
becomes significantly negative.</FONT></DIV>
<DIV><FONT size=2>You should refer to the table below to determine when a trend 
is deemed &#8220;significant&#8221;.</FONT></DIV>
<DIV><FONT size=2></FONT>&nbsp;</DIV>
<DIV><FONT size=2>For example, if the 14-period Slope has recently turned from 
negative to positive (i.e., crossed above zero),</FONT></DIV>
<DIV><FONT size=2>you may consider buying when r-squared crosses above the 0.27 
level.</FONT></DIV>
<DIV>&nbsp;</DIV>
<DIV><FONT size=2>To determine if the trend is statistically significant for a 
given x-period linear regression line, plot the</FONT></DIV>
<DIV><FONT size=2>r-squared indicator and refer to the following table.&nbsp; 
This table shows the values of r-squared required</FONT></DIV>
<DIV><FONT size=2>for 95% confidence level at various time periods.&nbsp; If the 
value is less than the critical values shown,</FONT></DIV>
<DIV><FONT size=2>you should assume that prices show no statistically 
significant trend.</FONT></DIV>
<DIV>&nbsp;</DIV>
<DIV><FONT size=2>Numberof Periods&nbsp;r-squaredCritical Value(95% 
confidence)<BR>5&nbsp;0.77<BR>10&nbsp;0.40<BR>14&nbsp;0.27<BR>20&nbsp;0.20<BR>25&nbsp;0.16<BR>30&nbsp;0.13<BR>50&nbsp;0.08<BR>60&nbsp;0.06<BR>120&nbsp;0.03<BR>You 
may even consider opening a short-term position opposite the prevailing trend 
when you observe</FONT></DIV>
<DIV><FONT size=2>the Slope rounding off at extreme levels.&nbsp; For example, 
if the Slope is at a relatively high level and begins</FONT></DIV>
<DIV><FONT size=2>to turn down, you may consider selling or opening a short 
position.</FONT></DIV>
<DIV>&nbsp;</DIV>
<DIV><FONT size=2>There are numerous ways to use the linear regression outputs 
of Slope and r-squared in trading systems.</FONT></DIV>
<DIV><FONT size=2>For more detailed coverage, refer to the book The New 
Technical Trader by Tushar Chande and Stanley Kroll.</FONT></DIV>
<DIV>&nbsp;</DIV>
<DIV><FONT size=2>See also R-squared.</FONT></DIV>
<DIV><FONT size=2><FONT 
size=2>---------------------------------------------------------------------------------</FONT></FONT></DIV>
<DIV><STRONG>Forecast Oscillator</STRONG></DIV>
<DIV>&nbsp;</DIV>
<DIV><FONT size=2>The Forecast Oscillator is an extension of the linear 
regression based indicators made popular</FONT></DIV>
<DIV><FONT size=2>by Tushar Chande.&nbsp; The Forecast Oscillator plots the 
percentage difference between the forecast price</FONT></DIV>
<DIV><FONT size=2>(generated by an x-period linear regression line) and the 
actual price.</FONT></DIV>
<DIV><FONT size=2>The oscillator is above zero when the forecast price is 
greater than the actual price.</FONT></DIV>
<DIV><FONT size=2>Conversely, it's less than zero if its below.</FONT></DIV>
<DIV><FONT size=2>In the rare case when the forecast price and the actual price 
are the same, the oscillator would plot zero.</FONT></DIV>
<DIV>&nbsp;</DIV>
<DIV><FONT size=2>For more on linear regression analysis</FONT></DIV>
<DIV><FONT size=2>See&nbsp;Linear Regression Trendline, Linear Regression 
Indicator, and Time Series Forecast.</FONT></DIV>
<DIV>&nbsp;</DIV>
<DIV><FONT size=2>===============================================</FONT></DIV>
<DIV>&nbsp;</DIV><FONT size=2>
<DIV>----- Original Message ----- 
<DIV>From: VonHef &lt;<A 
href="mailto:VonHef@xxxxxxxxxxxxx";>VonHef@xxxxxxxxxxxxx</A>&gt;</DIV>
<DIV>To: &lt;<A 
href="mailto:metastock@xxxxxxxxxxxxx";>metastock@xxxxxxxxxxxxx</A>&gt;</DIV>
<DIV>Sent: dinsdag 11 mei 1999 23:30</DIV>
<DIV>Subject: Re: Time Series Forecast Formula</DIV></DIV>
<DIV><BR></DIV>
<DIV>&gt; Hi Robert,<BR>&gt; &nbsp;What version of MetaStock are you using? The 
reason I ask<BR>&gt; is that 6.5 has the TSF built-in. Here is the format to use 
it:<BR>&gt; 
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; 
tsf( DATA ARRAY, PERIODS )<BR>&gt; Would this work for you?<BR>&gt; <BR>&gt; 
&nbsp; Best wishes,<BR>&gt; &nbsp;&nbsp;&nbsp;&nbsp;&nbsp; Adam Hefner.<BR>&gt; 
VonHef@xxxxxxxxxxxxx<BR>&gt; <BR>&gt; 
---------------------------------------<BR>&gt; ----- Original Message 
-----<BR>&gt; From: Robert Lambert &lt;<A 
href="mailto:lambertb1@xxxxxxxxx";>lambertb1@xxxxxxxxx</A>&gt;<BR>&gt; To: &lt;<A 
href="mailto:metastock@xxxxxxxxxxxxx";>metastock@xxxxxxxxxxxxx</A>&gt;<BR>&gt; 
Sent: Tuesday, May 11, 1999 9:07 AM<BR>&gt; Subject: Time Series Forecast 
Formula<BR>&gt; <BR>&gt; <BR>&gt; &gt; All:<BR>&gt; &gt;<BR>&gt; &gt; I would 
like to know if the following formula (taken from Equis<BR>&gt; &gt; website) is 
actually the formula for the Time Series Forecast, or a<BR>&gt; &gt; modified 
formula which is simply using the Time Series Forecast as part<BR>&gt; &gt; of 
it's computation.<BR>&gt; &gt;<BR>&gt; &gt; I'm asking because I'd like to setup 
a Time Series Forecast of an<BR>&gt; &gt; indicator as a crossover trigger, 
rather than use a moving average. So,<BR>&gt; &gt; if I plug an indicator into 
the below referenced formula( in place of<BR>&gt; &gt; the close value), will 
this particular formula actually give me the<BR>&gt; &gt; Time Series Forecast 
of the indicator, or will it give me something<BR>&gt; &gt; modified?<BR>&gt; 
&gt;<BR>&gt; &gt; Thanks in advance for feedback.<BR>&gt; &gt;<BR>&gt; 
&gt;<BR>&gt; &gt;<BR>&gt; &gt; The End Point Moving Average was introduced in 
the October 95 issue of<BR>&gt; &gt; Technical Analysis of Stocks &amp; 
Commodities in the article "The End<BR>&gt; &gt; Point Moving Average", by 
Patrick E. Lafferty.<BR>&gt; &gt; The exact formula for the End Point Moving 
average is as follows:<BR>&gt; &gt;<BR>&gt; &gt; ( 14 * Sum( Cum( 1 ) * C,14 ) - 
Sum( Cum( 1 ),14) * Sum( C,14) ) / (14<BR>&gt; &gt; * Sum( Pwr( Cum( 1 ),2),14 ) 
- Pwr( Sum( Cum( 1 ),14 ),2 ) ) * Cum( 1 )<BR>&gt; &gt; + (Mov(C,14,S) - Mov( 
Cum( 1 ),14,S) * (14 * Sum( Cum( 1 ) * C,14) -<BR>&gt; &gt; Sum( Cum( 1 ),14 ) * 
Sum( C,14) ) / (14 * Sum( Pwr( Cum( 1 ),2 ),14) -<BR>&gt; &gt; Pwr( Sum( Cum( 1 
),14 ),2 ) ) )<BR>&gt; &gt;<BR>&gt; &gt; The above formula plots the last value 
of a linear regression line of<BR>&gt; &gt; the previous 14 periods. The Time 
Series Forecast takes this value and<BR>&gt; &gt; the slope of the regression 
line to forecast the next day and then<BR>&gt; &gt; plots this forecasted price 
as today's value.<BR>&gt; &gt;<BR>&gt; &gt;<BR>&gt; &gt;<BR>&gt; &gt; 
_________________________________________________________<BR>&gt; &gt; Do You 
Yahoo!?<BR>&gt; &gt; Free instant messaging and more at <A 
href="http://messenger.yahoo.com";>http://messenger.yahoo.com</A><BR>&gt; 
<BR>&gt; <BR>&gt; <BR>&gt;</DIV>
<DIV>=====================================================</DIV>
<DIV>Long time in coming but here it is, sorry I am not in a cut and 
past<BR>environ so it took me some time to type it up: After an uptrend, when 
the<BR>mov crosses thru the trix expect a consolidation or emerging 
downtrend.<BR><BR><BR>PERIODS: WEEKLY<BR><BR>COLA: TRIX(3)<BR>COLB: 
REF(TRIX(3),-1)<BR>COLC: MOV(TRIX(3),8,TIMESERIES)<BR>COLD: 
REF(MOV(TRIX(3),8,TIMESERIES),-1)<BR>COLE: C<BR><BR>Filter 
enabled:yes<BR><BR>when(cola,&gt;,colc)and when(colb,&lt;,cold)and 
when(cola,&lt;,0)and<BR>when(cola,&gt;,-2)<BR><BR>On Mon, 19 Jan 1998 18:40:31 
-0500&nbsp;Mike writes:<BR>&gt;<BR>&gt;i tried it &amp; i don't get the same 
reults.<BR>&gt;please copy &amp; paste the formula &amp; i'll give it another 
shot.<BR>&gt;thanks for sharing.<BR>&gt;<BR>&gt;Mike</DIV>
<DIV>----------------------------------------------------------------------------</DIV>
<DIV>
<DIV>----- Original Message ----- 
<DIV>From: jeff </DIV>
<DIV>To: &lt;<A 
href="mailto:metastock-list@xxxxxxxxxxxxx";>metastock-list@xxxxxxxxxxxxx</A>&gt;</DIV>
<DIV>Sent: maandag 19 januari 1998 8:25</DIV>
<DIV>Subject: DJIA &amp; TRIX</DIV></DIV>
<DIV><BR></DIV>&gt; After reading an article on "playing trix" by Joe Luisi in 
S&amp;C June 97' I<BR>&gt; used his 3 time period MOV of&nbsp; Trix to look for 
advance signals. I have<BR>&gt; set up a template and an exploration in 
MS.&nbsp; All in all I am still<BR>&gt; playing with it but found that a 3 week 
trix seems to be predicting moves<BR>&gt; in the DOW by at least a week in 
advance.&nbsp; I load 25 weeks of data,<BR>&gt; create the trix indicator with a 
3 "time series" MOV on it.&nbsp; As the MOV<BR>&gt; crosses the trix indicator 
it gives signals, these were the most recent:<BR>&gt; 10/5/97 - sell<BR>&gt; 
11/7/97 - buy<BR>&gt; 12/12/97 - sell<BR>&gt; 1/9/98&nbsp; - 
buy</FONT></DIV></BODY></HTML>
</x-html>From ???@??? Sat May 15 19:57:28 1999
Received: from listserv.equis.com (204.246.137.2)
	by mail05.rapidsite.net (RS ver 1.0.2) with SMTP id 6453;
	Sat, 15 May 1999 19:53:10 -0400 (EDT)
Received: (from majordom@xxxxxxxxx)
	by listserv.equis.com (8.8.7/8.8.7) id RAA32130
	for metastock-outgoing; Sat, 15 May 1999 17:31:22 -0600
X-Authentication-Warning: listserv.equis.com: majordom set sender to owner-metastock@xxxxxxxxxxxxx using -f
Received: from freeze.metastock.com (freeze.metastock.com [204.246.137.5])
	by listserv.equis.com (8.8.7/8.8.7) with ESMTP id RAA32127
	for <metastock@xxxxxxxxxxxxxxxxxx>; Sat, 15 May 1999 17:31:19 -0600
Received: from smtp02.wxs.nl (smtp02.wxs.nl [195.121.6.60])
	by freeze.metastock.com (8.8.5/8.8.5) with ESMTP id RAA11874
	for <metastock@xxxxxxxxxxxxx>; Sat, 15 May 1999 17:41:52 -0600 (MDT)
Received: from escom ([195.121.181.56]) by smtp02.wxs.nl
          (Netscape Messaging Server 3.61)  with SMTP id AAB2E7F;
          Sun, 16 May 1999 01:29:11 +0200
Message-ID: <00ca01be9f2a$44cd4a60$LocalHost@xxxxx>
From: "A.J. Maas" <anthmaas@xxxxxx>
To: "Metastock-List" <metastock@xxxxxxxxxxxxx>
Subject: Re: Time Series Forecast Formula (part2)
Date: Sun, 16 May 1999 00:46:30 +0200
Organization: Ms-IRB
MIME-Version: 1.0
Content-Type: multipart/mixed;
	boundary="----=_NextPart_000_005A_01BE9F35.89830220"
X-Priority: 3
X-MSMail-Priority: Normal
X-Mailer: Microsoft Outlook Express 5.00.2014.211
X-MimeOLE: Produced By Microsoft MimeOLE V5.00.2014.211
Sender: owner-metastock@xxxxxxxxxxxxx
Precedence: bulk
Reply-To: metastock@xxxxxxxxxxxxx
X-Loop-Detect: 1
X-UIDL: 93d16fb254dea043b5e4d88bf81b1e32.46

The TSF.gif (see explanation in mail-part1)

Regards,
Ton Maas
ms-irb@xxxxxxxxxxxxx
Dismiss the ".nospam" bit (including the dot) when replying.


Attachment Converted: "c:\eudora\attach\TSF.gif"