eSignal EFS code
- Thread starter Magna
- Start date
I went and did a search on esignal trying to remember who wrote the script. It is written by one of the better esignal efs official third party programmers. Alex Montenegro
I was frustrated because I don't like posting things that waste other people's time and that is what I felt like I did when it has worked fine for me for a long time now. The same issue discussed in this thread was also pointed out to the programmer in a thread on esignal bulletin board. Since I only use it in realtime I haven't seen the problem mention by the ET members. However going back and looking using historical data I see the problem Easy and Magna are seeing.
Here is a quote from the programmer on how to determine if the script is working correctly. Made sense to me after reading it.
Quote
"First of all to verify if the efs is plotting the correct values run it on a chart that has the same symbol as the one in the efs. You will very likely see that the values returned by the efs match exactly those of the symbol in the price window.
Then try changing the Time Templates to 8:30-15:00 (not 15:15 as you have them) in both your charts and you should also see the efs return the correct values.
Essentially, the data sets for both symbols need to be identical in the time series for the efs to stay in sync hence you need to use the shorter of the trading sessions to do that otherwise the plots go out of sync.
Following is a visual example to illustrate the issue. Each letter is a different 15 minute bar with A starting at 08:30 CT. The last bar on $INDU (top series) will be V at 14:45. However your Time Template includes the 15:00 bar also (shown as x in the lower series).
_ABCDEFGHIJKLMNOPQRSTUV
ABCDEFGHIJKLMNOPQRSTUVx
The efs aligns the plot starting from the last bar going backwards. However it does not align by time but by relative bar index hence going backwards past the x the two plots are out of sync. They will be correctly aligned in real time on the current bar though as data from a new day gets added
_ABCDEFGHIJKLMNOPQRSTUVABC
ABCDEFGHIJKLMNOPQRSTUVxABC
If you were to use an 8:30-15:00 Time Template then the 15:00 bar (x) that is causing this issue would be omitted and the historical values should realign."
Hopefully this clears things up some. At least I hope it doesn't make it worst. :-(
Bsulli
Also a link to the thread over on esignal.
http://forum.esignalcentral.com/showthread.php?threadid=7085&highlight=tickextreme.efs
I was frustrated because I don't like posting things that waste other people's time and that is what I felt like I did when it has worked fine for me for a long time now. The same issue discussed in this thread was also pointed out to the programmer in a thread on esignal bulletin board. Since I only use it in realtime I haven't seen the problem mention by the ET members. However going back and looking using historical data I see the problem Easy and Magna are seeing.
Here is a quote from the programmer on how to determine if the script is working correctly. Made sense to me after reading it.
Quote
"First of all to verify if the efs is plotting the correct values run it on a chart that has the same symbol as the one in the efs. You will very likely see that the values returned by the efs match exactly those of the symbol in the price window.
Then try changing the Time Templates to 8:30-15:00 (not 15:15 as you have them) in both your charts and you should also see the efs return the correct values.
Essentially, the data sets for both symbols need to be identical in the time series for the efs to stay in sync hence you need to use the shorter of the trading sessions to do that otherwise the plots go out of sync.
Following is a visual example to illustrate the issue. Each letter is a different 15 minute bar with A starting at 08:30 CT. The last bar on $INDU (top series) will be V at 14:45. However your Time Template includes the 15:00 bar also (shown as x in the lower series).
_ABCDEFGHIJKLMNOPQRSTUV
ABCDEFGHIJKLMNOPQRSTUVx
The efs aligns the plot starting from the last bar going backwards. However it does not align by time but by relative bar index hence going backwards past the x the two plots are out of sync. They will be correctly aligned in real time on the current bar though as data from a new day gets added
_ABCDEFGHIJKLMNOPQRSTUVABC
ABCDEFGHIJKLMNOPQRSTUVxABC
If you were to use an 8:30-15:00 Time Template then the 15:00 bar (x) that is causing this issue would be omitted and the historical values should realign."
Hopefully this clears things up some. At least I hope it doesn't make it worst. :-(
Bsulli
Also a link to the thread over on esignal.
http://forum.esignalcentral.com/showthread.php?threadid=7085&highlight=tickextreme.efs
Quote from easyrider:
They seem to be in synch this morning.
I've bee using it for months in realtime and also run Ensign on the same pc with a $tick chart . It and the esignal tickextreme.efs are always in sync during the trading day.
Personally I have a $trin chart with the tickextreme.efs as the study on it for my use. Saves me some monitor space for other charts. I just watch for the study bars to turn red and take the reversal.
Good luck and good trading!
Bsulli
Magna
Administrator
Nice to see no 2hr offset. However, since I'm still keeping an actual $TICK chart just below for reference, I notice that the study often doesn't properly display. That is, bars don't quite match up with the true $TICK chart, with adjacent ones often showing similar highs or lows (when, in fact, they shouldn't be the same), and it doesn't self-correct no matter how long I wait. If I manually refresh whatever chart I have the TickExtreme.efs "attached" to then the study will correct and match the $TICK chart, but not until. Obviously whatever information the study is processing is different than the information feeding a regular $TICK chart.Quote from easyrider:
They seem to be in synch this morning.
Magna, I think the problem in your charts is the time frame. It will be fine for the current day, but when you look back at previous charts, the $TICK will become more and more offset (by 15 minutes each day) from the main chart. For instance, in your ES chart, the $TICK value that will correspond to the 16:15 bar will be the $TICK from the NYSE at 15:55. The tick for the ES at 15:55 will be that from the NYSE at 15:40.
You can probably get around this problem by simply adding a line into the EFS... something like:
if(getHour() < 16) {
tick = close("$TICK, 5");
}
else {
tick = 0;
}
That statement would work in the small tick study I posted. Something similar should work in any other tick studies. If your time frame also includes the pre-market, more than just these lines will be needed to put the study in synch with the main chart.
Hope this helps.
You can probably get around this problem by simply adding a line into the EFS... something like:
if(getHour() < 16) {
tick = close("$TICK, 5");
}
else {
tick = 0;
}
That statement would work in the small tick study I posted. Something similar should work in any other tick studies. If your time frame also includes the pre-market, more than just these lines will be needed to put the study in synch with the main chart.
Hope this helps.