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

RE: [amibroker] Re: Time Series Forecast/Forecast Oscillator



PureBytes Links

Trading Reference Links


Steve:thanks for 
sharing the idea.  The chart you said you included does not display for 
me....Outlook says it has html code and it can not be displayed.  Can you 
capture a gif and send that to the group??
<SPAN 
class=060032817-22042002> 
<SPAN 
class=060032817-22042002>Thanks,
<SPAN 
class=060032817-22042002> 
Ken 

<FONT face=Tahoma 
size=2>-----Original Message-----From: Steve Karnish 
[mailto:kernish@xxxx]Sent: Monday, April 22, 2002 12:23 
PMTo: amibroker@xxxxxxxxxxxxxxxSubject: Re: [amibroker] 
Re: Time Series Forecast/Forecast Oscillator
CS & list,
 
I'm not up to speed on AB code...yet.  Here's 
a TSF approach that can work well.  Attached is a chart of KLAC, showing 
triggers on the Forecast Oscillator (same indicator).  The Forecast 
Oscillator is a TSF set to 13 periods (in this example).  In this case, 
whenever price moves more than 11% (in the opposite direction of your position) 
from the projected linear regression, you reverse your position (all signals are 
calculated EOD, and position are put on the following opening) .  Since 
1/1/2000, this has produced results, in KLAC, of 28-9 with a return of $159(no 
slippage or commissions).  Just thought I'd toss in an approach that, at 
times, can extract a few pennies.
 
Take care,
 
Steve
<A 
href="">www.cedarcreektrading.com
 
 
<BLOCKQUOTE 
>
----- Original Message ----- 
<DIV 
>From: 
CS 
To: <A 
href="" 
title=amibroker@xxxxxxxxxxxxxxx>amibroker@xxxxxxxxxxxxxxx 
Sent: Monday, April 22, 2002 8:24 
AM
Subject: Re: [amibroker] Re: Time Series 
Forecast

"The Time Series Forecast indicator is based on the trend of 
a security's price over a specified time period.  The trend is determined 
by calculating a linear regression trendline using the "least squares fit" 
method.  The least squares fit technique fits a trendline to the data in 
the chart by minimizing the distance between the data points and the linear 
regression trendline.  Any point along the Time Series Forecast is 
equal to the ending value of a Linear Regression trendline plus its 
slope.  For example, the ending value of a Linear Regression trendline 
(plus its slope) that covers 10 days will have the same value as a 10-dayTime 
Series Forecast.  This differs slightly from the Linear Regression 
indicator (see Linear Regression Indicator) in that the Linear Regression 
indicator does not add the slope to the ending value of the regression 
line.  This makes the TSF a bit more responsive to short term price 
changes.  If you plot the TSF and the Linear Regression indicator 
side-by-side, you'll notice that the TSF hugs the prices more closely than the 
Linear Regression indicator.
 
Rather than plotting a straight Linear Regression trendline, 
the Time Series Forecast indicator plots the ending values of multiple Linear 
Regression trendlines.  The resulting Time Series Forecast indicatoris 
sometimes referred to as a "moving linear regression" study or a "regression 
oscillator.""
 
I use it as a type of moving average or smoother of other 
indicators. Sometimes.
 
-CS

<BLOCKQUOTE 
>
----- Original Message ----- 
<DIV 
>From: 
Cathy 
Finnegan 
To: <A 
href="" 
title=amibroker@xxxxxxxxxxxxxxx>amibroker@xxxxxxxxxxxxxxx 
Sent: Sunday, April 21, 2002 5:54 
PM
Subject: RE: [amibroker] Re: Time 
Series Forecast

<FONT color=#0000ff face=Arial 
size=2>This looks interesting!  I just plotted it,,,  What's 
it supposed to ..er..  indicate..  show..  
do??
<FONT color=#0000ff face=Arial 
size=2> 
<FONT color=#0000ff face=Arial 
size=2>cathy

<FONT face=Tahoma 
size=2>-----Original Message-----From: CS 
[mailto:csaxe@xxxx]Sent: Sunday, April 21, 2002 8:42 
PMTo: <A 
href="">amibroker@xxxxxxxxxxxxxxx; <A 
href="">pacific5_au@xxxxSubject: 
Re: [amibroker] Re: Time Series Forecast

See if this gets through.
 

T1=7; //Periods
Y=C; //array
<FONT 
color=#0000ff>TSF=LinRegSlope(Y,T1)+((T1*Sum(Cum(1)*Y,T1)-Sum(Cum(1),T1)*Sum(Y,T1))/(T1*Sum((Cum(1)^2),T1)-
<FONT 
color=#0000ff>(Sum(Cum(1),T1)^2))*Cum(1)+(MA(Y,T1)-MA(Cum(1),T1)*(T1*Sum(Cum(1)*Y,T1)-
<FONT 
color=#0000ff>Sum(Cum(1),T1)*Sum(Y,T1))/(T1*Sum((Cum(1)^2),T1)-(Sum(Cum(1),T1)^2))));
GraphXSpace=2;
Graph0=TSF;
Graph1=Y;
Title="Time Series Forecast 
Periods="+WriteVal(T1);
-CS
<BLOCKQUOTE 
>
----- Original Message ----- 
<DIV 
>From: 
<A href="" 
title=pacific5_au@xxxx>pacific5_au 
To: <A 
href="" 
title=amibroker@xxxxxxxxxxxxxxx>amibroker@xxxxxxxxxxxxxxx 
Sent: Sunday, April 21, 20025:08 
PM
Subject: [amibroker] Re: Time 
Series Forecast
Hi CSThanks for your post. Unfortunatelythe 
file is empty 0 kb when down loaded. Could you plse re-up the file 
as a zip file or just post the code text. (a Yahoo 
situation.)Many thankspacific--- In <A 
href="">amibroker@xxxx..., "CS" <<A 
href="">csaxe@xxxx...> wrote:> 
-CS>   ----- Original Message ----- 
>   From: Owen Davies >   To: <A 
href="">amibroker@xxxx... >  Sent: 
Sunday, April 21, 2002 8:25 AM>   Subject: [amibroker] 
Time Series Forecast> > >   Has anyone 
coded this Metastock indicator for AmiBroker?>   
According to the Equis Help description, it's the end point 
of>   the linear regression line plus the slope.  
Not something I'm>   competent enough to produce for 
AB, alas.> >   Thanks.> 
>   Owen Davies> > >   
Your use of Yahoo! Groups is subject to the Yahoo! Terms of 
Service.Your use of Yahoo! Groups is 
subject to the Yahoo!Terms 
of Service. Your use of Yahoo! 
Groups is subject to the <A 
href="">Yahoo! Terms of Service. 
Your use of Yahoo! Groups is subject to the <A 
href="">Yahoo! Terms of Service. 
Your use of Yahoo! Groups is subject to the <A 
href="">Yahoo! Terms of Service. 
Your 
use of Yahoo! Groups is subject to the <A 
href="">Yahoo! Terms of Service.