New Scripts

[mungo]

Renko Kagi implementation

Chris,
You did a great job on these scripts. Thanks for putting your time into producing such original ideas. I now don't have to have two charts up , just a candlestick with the Kagi implementation.

cheers Kelvin

[ckryza]

A new version of the Renko script has been uploaded to the Specialty Scripts group that fixes a problem with the Renko calcs in the original script.

Link to File Share:

http://share.esignal.com/groupconten...rt&groupid=114

Also, I added some new SineWeighted average scripts to the Miscellaneous group over the last month or so.

Chris

[ckryza]

New version of the EFS Help File (Windows CHM and standard HLP) has been uploaded to the Specialty Script group.

Link to File Share:

http://share.esignal.com/groupconten...le&groupid=114

Chris

[ckryza]

Going forward, the current versions of both the dsFunctions and dsUtilities libraries (as well as the documentation) can be found in my Specialty Scripts group here:

http://share.esignal.com/groupconten...es&groupid=114

Chris

[richard15]

existing filters

Chris:

I am a retired physicist evaluating eSignal for purchase.

In your post of 04-07-2003 12:50 PM you mentioned at least five filters that could potentially remove noise from data without the disadvantages like time displacement and curve flattening, which are inherent in moving or rolling averages:

1) Fourier,
2) Lomb,
3) MESA,
4) Wavelets,
5) Kalman.

A) Please recommend off the shelf products that implement each of these as filters to remove noise from eSignal graphs.

I am unfamiliar with all these filters, that is, how they work, except for Fourier. Relative to Fourier, I found forward Fourier transforms like at Metastock but found no products incorporating the inverse transforms, let alone any that completed the filtering of noise from the data.

B) Please recommend off the shelf products that could display the slope of the smoothed output overlaid on an eSignal graph.

C) Are any of them unsuitable for using their output as inputs to other studies, for example MACD? Or do all your recommended products work as inputs to other studies?

Originally posted by ckryza
Hi Joshua:

A typical use for Wavelets would be in time series analysis (along with Fourier transforms and MESA). They can be used to break down a time series into it's frequency components and they can also be used to filter out certain frequencies in the time series (e.g., remove the noise from the data). The wavelet algorithms, on average, tend to be faster and more efficient (in terms of CPU usage) than Fourier, Lomb and/or MESA algorithms which makes them popular for real-time applications. (However, the Lomb and MESA algorithms are more robust (in my opinion) for cycle analysis work.)

A direct application of Wavelets would be to use them to create a filtered signal as mentioned above. The filtered signal could be used in the same fashion as a Kalman filter or an adaptive moving average. The idea being that the filtered signal would be less sensitive to minor fluctuations in the data and, therefore, less prone to whipsaw. Another application would be in the area of neural networks.... which often use wavelets as a mechansim for pre-processing the raw time series data.

Chris

[ckryza]

Jurik Function Library

I put together an EFS2 function library that encapsulates the 4 EFS1 Jurik functions (e.g., JMA, RSX, DMX and VEL). This allows you to use the Jurik functions in conjunction with other EFS2 series as well as to implement the Jurik studies on different bar intervals.

The new library as well as some example scripts are in a Zip file which can be found here:

http://share.esignal.com/groupconten...es&groupid=114

Just download, unzip and copy the efsLib file into your FunctionLibrary directory and copy the example EFS scripts into any eSignal Formula Directory. Requires eSignal Version 7.91 (Build 732) or newer to operate.

(Note: You must be a registered Jurik user for the library to work.)

Chris

[ckryza]

I added a new MACD function to the EFS2 Jurik library discussed in the message below. It allows you to create a MACD indicator based on JMA's with one function call.

The updated library as well as example scripts of all of the functions can be found here:

http://share.esignal.com/groupconten...es&groupid=114

Just download, unzip and copy the efsLib file into your FunctionLibrary directory and copy the example EFS scripts into any eSignal Formula Directory. Requires eSignal Version 7.91 (Build 732) or newer to operate.

(Note: You must be a registered Jurik user for the library to work.)

Chris

[drizzi]

Chris

The new MACD doesn't plot. The rest work great.

Doug

[ckryza]

Works just fine on my end. Are you sure you copied the new version of the efsLib over the original version in the FunctionLibrary folder?

Chris

[drizzi]

Hey Chris

Yes, I double checked and repeated the download procedure. All the EFS' work except for the MACD. The study window/pane opens but there is no plot. Very strange. I'm running a black background so it isn't a color issue and I looked at the script and don't see anything obvious.

Doug

[drizzi]

Disregard my last post, I got it working. Thanks

[ckryza]

A new version of the dsUtilities library (V1.25) has been uploaded to the File Share. It implements a few bug fixes and takes advantage of some functionality that has been added to eSignal since the library was first released.

Note that this version of the library requires eSignal 7.91 (Build 732) or newer in order to operate.

Link to File Share:

http://share.esignal.com/groupconten...es&groupid=114

Chris

[ckryza]

A new version of the dsUtilities library (v1.27) has been uploaded to the File Share. It contains a new function (dsPlayWAV()) that allows you to play a specified WAV file multiple times in succession.

A corresponding DLL file is required and is also available in the File Share. Just save the DLL directly into your eSignal Program Directory ( generally c:\Program Files\eSignal ) prior to using the new dsPlayWAV() function.

Documentation has also been updated.

Note that this version of the library requires eSignal 7.91 (Build 732) or newer to operate.

Link to File Share:

http://share.esignal.com/groupconten...es&groupid=114

Chris

[adryden]

murrey math octaves efs

i really like the murrey math octave study and was wondering if it was possible to display the octave labels (1/8, 2/8, etc).

thank you,
angela

[ckryza]

I placed a modified version in File Share that displays the 8th levels.

http://share.esignal.com/groupconten...es&groupid=114

Chris