Issues with Ergotic TSI and Bressert-DS

[mattgundersen]

Hello All,

I've read some posts over on the eSignal/Yahoo Groups about the Ergotic TSI and Bressert DS EFS formulas not updating properly.

What I'd like to do is discuss these formulas in this thread with the users seeing the problem and TSSupport to better understand the issue.

My hope is that we can all get an understanding as to what the problem is and get this corrected.

On that note, the people seeing the problem with the Ergotic TSI and Bressert DS formulas, could you please post what the problem is? Perhaps put some screenshots on the fileshare and post the URL in here, or attach them to a post here?

Side note: If you post the image to the fileshare, you can use "vB Code" to make the image appear inline using the [ img ] tag. Click here http://forum.esignalcentral.com/misc...bbcode#buttons to learn more.

Thanks. Matt Gundersen

[TS Support]

Latest Bressert-DS version

Matt, here is the code for Bressert-DS study.

/**********************************************
Description	: This Indicator plots DSS Bressert
Provided By	: TS Support, LLC for eSignal. (c) Copyright 2002 
*********************************************/

function preMain()
{
    setStudyTitle("DSS Bressert");
    setCursorLabelName("DSS", 0);
    setCursorLabelName("Trigger", 1);
    setDefaultBarFgColor(Color.brown, 0);
    setDefaultBarFgColor(Color.red, 1);
    addBand(20, PS_SOLID, 1, Color.black);
    addBand(80, PS_SOLID, 1, Color.black);
}

var EMA_1 = 0;
var EMA1_1 = 0;
var EMA2_1 = 0;
vAA = new Array();

function main(pds,slw,triggerLen,emalen){
	if (pds == null) 
		pds = 10;
	if (slw == null) 
		slw = 3;
	if (triggerLen == null) 
		triggerLen = 5;
	if (emalen == null) 
		emalen = 9;
	
	var vHigh = getValue("High",0,-pds);
	var vLow = getValue("Low",0,-pds);
	var dClose = getValue("Close");
	if(vHigh == null || vLow == null || dClose == null)
		return;
   	
   	var K = 2 / (emalen + 1);
    	var K2 = 2 / (triggerLen + 1);
    	var aa = 0;
    	var dss = 0;
	var High = 0;
	var Low = low();
	var aHigh = 0;
	var aLow = 0;
	var EMA1 = 0;
	var EMA = 0;
	
	for (i = 0; i < pds; i++){
        	High = Math.max(High,vHigh[i]);
        	Low =  Math.min(Low,vLow[i]);
	}
   	if((High - Low) != 0)
   		EMA = K * ( (dClose - Low) / (High - Low) ) + (1 - K) * EMA_1;
   	else
   		EMA = EMA_1;
	if (getBarState() == BARSTATE_NEWBAR)
		EMA_1 = EMA;
	aa = EMA * 100;
	if (getBarState() == BARSTATE_NEWBAR){
		for(i = pds - 1; i > 0; i--)
			vAA[i] = vAA[i - 1];
		vAA[0] = aa;
	}
   	for (i = 0; i < pds; i++){
   		if(i == 0)
   			aLow = vAA[0];
        	if (aHigh < vAA[i])
        		aHigh = vAA[i];
        	if (aLow > vAA[i])
        		aLow = vAA[i];
	}
	if (aHigh - aLow != 0)
		EMA1 = K * ( (aa - aLow) / (aHigh - aLow) ) + (1 - K) * EMA1_1;
	else
		EMA1 = EMA1_1;
	if (getBarState() == BARSTATE_NEWBAR)
		EMA1_1 = EMA1;
	dss = EMA1 * 100;
	var EMA2 = K2 * dss + (1 - K2) * EMA2_1;
	if (getBarState() == BARSTATE_NEWBAR)
		EMA2_1 = EMA2;
	return new Array(dss,EMA2);
}
/*************************************************/

[mattgundersen]

Posting Code

Hey Robert, thanks for the prompt response!

BTW, when anyone posts code, if you put

[ code ]
your code goes here
[ /code ]

Then the code will be formatted nicely. Note: I added spaces inside the brackets so you could see the tagnames. You don't need to put spaces inside the [ ] (brackets)

[mattgundersen]

Ergotic

Can anyone tell me what the issues are with Ergotic TSI?

[TS Support]

Test

Testing...

function main(pds,slw,triggerLen,emalen){
	if (pds == null) 
		pds = 10;
                 return pds;
}

Matt, your avatar is delightful

[gspiker]

Example

This is an example of the problem (lets hope I can inline this correctly ;-)



This is a 1 min IBM chart. The lower studies are both ergotic_TSI that were started at about the same time. The price study is two kama studies, also started at about the same time.

After running for a while, I refreshed one of the KAMA and one of the lower Ergotic TSI. Notice the differences between the two. The refreshed ones are the correct values/look, the ones run in RT are different. The parts of the charts that look the same for both is where I started both fomrula's.

The thing you can't tell here, (but could surmise I bet), is that when running in RT, the formula line being built seems to react to the price quicker (by this I mean that the current bar price seems to have more influence) than it should.

People have been mentioning that they think it is an EMA priming issue, but I don't think so. The MACD formula I sent you displays the same problems and uses an EMA primed by an SMA for both the fast and slow lines.

Also, the refreshed formula very quickly catches up to nearly matching the values for the unfreshed formula...also indicating that too much weight is being placed on the most current prices.

[gspiker]

Same charts but with more data (after refresh of lower and 1 KAMA).

[Doji3333]

TSI Still has a problem...

Matt and others,

I was just talking to a client yesterday and he stated the TSI problem still exists in the NEW BETA. If anyone can track this down, it would be a big help..

Brad

[vladar]

Same issue with new efs files posted by esignal

I just put two new studies on a chart that were provided/posted here on esignal last week:

The Blau Candlestick Indicator and Demand Index.
Both have the same problem as the ergotic: if the chart is not refreshed manually, the indicators do not draw the line properly as time progresses.

I would post a chart if you think it's useful, but it's just the same issue as the ergotic.

I am running this on the new beta release.

[wavetrader]

Re: Ergotic

Originally posted by mattgundersen
Can anyone tell me what the issues are with Ergotic TSI?

It is still not updating properly. I've got Build 579 already. I have to constantly keep reloading the indicator to make sure I'm getting the proper values.

Here is an example of Ergotic TSI-2 without refresh.

[wavetrader]

Re: Ergotic

Originally posted by mattgundersen
Can anyone tell me what the issues are with Ergotic TSI?

(continuation of previous e-mail)

Here's the Ergotic TSI-2 3 seconds later after a refresh:

Can somebody please fix this so that it automatically updates. It's a bit annoying having to refresh it all the time.

[eSignal_JasonK]

Hello wavetrader,

There was a corrected version of this in File share. Please give this version a try.

Ergotic_TSIMAG.efs

[wavetrader]

Originally posted by JasonK
Hello wavetrader,

There was a corrected version of this in File share. Please give this version a try.

Ergotic_TSIMAG.efs

Hi Jason,

Thanks for this. It is the corrected version of Ergotic TSI.
But how about the corrected version of Ergotic TSI-2?
I use TSI-2 more than the original as it gives more precise entries for me.

Also, where do we get the updated versions of all efs files?

Thanks.

[wavetrader]

Jason,

Since I did not get a response if there is a corrected version of the
Ergotic TSI-2 indicator, I went ahead and tweaked it myself.

It seems to be working fine now, and is updating correctly.

I'm posting it here for everybody's use.

Matt, can you please review it and if there are any corrections, please let me know asap, as I use this indicator quite a bit.

Thanks.
wavetrader

[jouster]

Explanation of Ergotic TSI-2 parameters?

Hello Wavetrader,

I found your post of a couple years ago regarding the Ergotic TSI-2 indicator. I've been finding it very helpful in trading but have been curious if there's an explanation anywhere for the eight parameters (r1,s1,u1, etc.) which allow adjustment in Edit Studies. I've spent time tweaking them by trial and error but as they aren't labeled clearly its not easy to think through how I might take advantage of multiple adjustments. If you or anyone else has seen an expalnation of the contruction of the formula I'll appreciate the information. Thanks.

Jouster