Indicator only updates dynamically ?

[Desert Dan]

Can someone kindly enlighten me as to why my example indicator (below) will not draw a line for the entire length of chart history but instead only start from current time going forwards (dynamic only, from loading time)?

Apologies in advance if it is blindingly obvious....

1000 Thanks, DD

/*************************
.DD Test
***************************/

function preMain() {
setPriceStudy(false);
setStudyTitle("DD Test")
setCursorLabelName("DD Test",0);
setDefaultBarFgColor(Color.blue,0)
setShowTitleParameters( true );
}
var LRs = null;

function main() {

if (LRs == null) LRs = middleLinearReg(21, 2,close());

return sma(8, LRs);
}

[Desert Dan]

No reply still? Must be a hard one....

Basically the Lin Reg indicator alone works perfectly across the whole chart history, as too does the SMA. But when you use the first as an input to the second you get no resulting line.

Is this a problem with one being a 'series' and perhaps not being called up as such?

Please help if anyone has any clues. I am new to this language and frankly am confused with things such as getValue, getSeries etc.

Thanks in advance.
DD

[waynecd]

Hi,

The LR indicator only plots the defined length as can be seen below for the following efs script:

function preMain() {
setPriceStudy(false);
setStudyTitle("DD Test")
setCursorLabelName("DD Test",0);
setDefaultBarFgColor(Color.blue,0)
setShowTitleParameters( true );
}
var LRs = null;
function main() { 
    return middleLinearReg(21, 2,close());
} 

Wayne

[waynecd]

Try this efs to which I added a SMA(8) to match your script.

Wayne

/*********************************
//http://forum.esignal.com/showthread.php?s=&threadid=30196&highlight=SEB+LinRegSlope
Provided By:    eSignal (Copyright c eSignal), a division of Interactive Data
    Corporation. 2009. All rights reserved. This sample eSignal
    Formula Script (EFS) is for educational purposes only and may be
    modified and saved under a new file name.  eSignal is not responsible
    for the functionality once modified.  eSignal reserves the right
    to modify and overwrite this EFS file with each new release.
Description:          This indicator plots the slope of linear regression in oscilator-like
    manner. This is a part of Standard Error Bands study.
Version:            1.0  03/19/2009
Formula Parameters:                     Default:
    Length                              21
Notes:
    Linear Regression is a concept also know as the "least squares method" or
    "best fit". Linear Regression attempts to fit a straight line between several
    data points in such a way that distance between each data point and the line is
    minimized.
    "Standard Error Bands" Jon Anderson, Stocks&Commodities Magazine, Traders Tips, 09/1996
**********************************/
//20110524 added SMA of LR
var fpArray = new Array();
var bInit = false;
function preMain() {
    setStudyTitle("SEB LinRegSlope");
    setCursorLabelName("Slope", 0);
    setDefaultBarFgColor(Color.red, 0);
    addBand(0, PS_SOLID, 1, Color.cyan);

    var x=0;
    fpArray[x] = new FunctionParameter("Length", FunctionParameter.NUMBER);
    with(fpArray[x++]){
        setLowerLimit(1);
        setDefault(21);
    }
}
var xSL = null;
var SMA8;
function main(Length) {

var nBarState = getBarState();
var nSL = 0;
    if (nBarState == BARSTATE_ALLBARS) {
        if (Length == null) Length = 21;
    }
    if (bInit == false) {
        xSL = efsInternal("calc", Length);
        SMA8 = sma(8, xSL);
        SMA8 = getSeries(SMA8);
        bInit = true;
    }
    nSL = xSL.getValue(0);
    if (nSL == null) return;
    return SMA8;//nSL;
}
var xClose = null;
var xInit = false;

function calc(Length) {
var SL = 0;
var SumBars = Length * (Length - 1) * 0.5;
var SumSqrBars = (Length - 1) * Length * (2 * Length - 1) / 6;
var Sum1 = 0;
var SumY = 0;
var i = 0;

    if(xInit==false){
        xClose = close();
        xInit = true;
    }
    for (i = 0; i < Length; i++) {
        Sum1 += i * xClose.getValue(-i);
        SumY += xClose.getValue(-i);
    }
    var Sum2 = SumBars * SumY;
    var Num1 = Length * Sum1 - Sum2;
    var Num2 = SumBars * SumBars - Length * SumSqrBars;
    SL = Num1 / Num2;
    return SL;
} 

[Desert Dan]

Thanks for your reply Wayne.
I get an error when I run your code:
"SyntaxCheck, line 26: Formula file not found:"

Actually I must apologise. I realised earlier that it (middleLinearReg) prints only as far back as the defined length and then found the code below instead, which I used, and which did work across the whole range. This is what I should have been referring to in second post and got the two confused after trying so many variations.

function main() {
var n = 21;
var sum = 0;
var i = 0;
var LRs = 0;

for(i = n; i > 0; i--)
sum += (i - (n + 1) / 3) * close(i - n)
LRs = 6 / (n * (n + 1)) * sum;
return LRs

Now here the LRs works perfectly across the chart.

But how to take SMA of this?

I tried sma(8,"LRs") but this just returns sma(8) of close.
How does one get it to refer to LRs as the input?

Note, lengths here are just for example, nothing specific.

Thanks.

[waynecd]

The study I posted works fine in ver 10.6 latest release. Line 26 reads:

setCursorLabelName("Slope", 0); 

so it has nothing to do with calling a file. Not sure why you get an error, maybe there is another study loaded on the chart?

I don't use ver 11 since it is not ready for prime time yet (has issues with studies that use tick intervals).

In any case the following works in 10.6 and uses your formula:

var bInit = false;//limits execution to once upon script loading
var nLRs, nSMA;
function main() { 
    if(!bInit){//if bInit is false
        nLRs = efsInternal("LRcalc");//creates LR series - lookup "efsInternal" in the knowledgebase
        nSMA = sma(8, nLRs);//uses "nLRs" series as the souce of data
        nSMA = getSeries(nSMA);
        bInit = true;
    }
    return nSMA;
}
function LRcalc(){//separate function so it creates a series for use in the sma(...) calc
    var n = 21;
    var sum = 0;
    var i = 0;
    var LRs = 0;

    for(i = n; i > 0; i--)
    sum += (i - (n + 1) / 3) * close(i - n)
    LRs = 6 / (n * (n + 1)) * sum; 
    return LRs
} 

This is the chart displaying both studies:

Wayne

[Desert Dan]

Out of interest Wayne, what happens if you swap the two indicators around - taking LR of the SMA ?

I tried the following but get nothing. I imagine that again it is a "series"issue. Anything obviously wrong here?

Thanks, DD.


var bInit = false;//limits execution to once upon script loading
var nLRs, nSMA;

function main() {

if(!bInit){//if bInit is false
nSMA = sma(8, "close");
nSMA = getSeries(nSMA);
nLRs = efsInternal("LRcalc");//creates LR series - lookup "efsInternal" in the knowledgebase
bInit = true;
}
return nLRs;
}
function LRcalc(){//separate function so it creates a series
var n = 21;
var sum = 0;
var i = 0;
var LRs = 0;

for(i = n; i > 0; i--)
sum += (i - (n + 1) / 3) * nSMA(i - n)
LRs = 6 / (n * (n + 1)) * sum;
return LRs
}

[waynecd]

In this particular case it seems both (returning the nLRs or the nSMA) methods produce the same plot.

var bInit = false;//limits execution to once upon script loading
var nLRs, nSMA;

function main() { 

    if(!bInit){//if bInit is false
        nSMA = sma(8, "close");
        nSMA = getSeries(nSMA);
        nLRs = efsInternal("LRcalc",nSMA);//creates LR series - lookup "efsInternal" in the knowledgebase
        bInit = true;
    }
    return nLRs.getValue(0);
}
function LRcalc(sSMA){//separate function so it creates a series 
    var n = 21;
    var sum = 0;
    var i = 0;
    var LRs = 0;

    for(i = n; i > 0; i--){
        sum += (i - (n + 1) / 3) * sSMA.getValue(i - n);
    }  
    LRs = 6 / (n * (n + 1)) * sum; 
    return LRs;
} 

This is the same script with a small coding difference:

var bInit = false;//limits execution to once upon script loading
var nLRs, nSMA;

function main() { 

    if(!bInit){//if bInit is false
        nSMA = sma(8, "close");
        nSMA = getSeries(nSMA);
        nLRs = getSeries(efsInternal("LRcalc",nSMA));//creates LR series - lookup "efsInternal" in the knowledgebase
        bInit = true;
    }
    return nLRs;//.getValue(0);
}
function LRcalc(sSMA){//separate function so it creates a series 
    var n = 21;
    var sum = 0;
    var i = 0;
    var LRs = 0;

    for(i = n; i > 0; i--){
        sum += (i - (n + 1) / 3) * sSMA.getValue(i - n);
    }  
    LRs = 6 / (n * (n + 1)) * sum; 
    return LRs;
} 

Wayne

[Desert Dan]

Many thanks again Wayne. Looks so simple when you know what you are doing.
I see I have a lot to learn here about the intricacies of series and functions but you have gone a long way in helping me already by showing me what you have in these few examples. Much appreciated....

Cheers.
DD

[waynecd]

yw

Wayne