team17
team17
Not a bug? This kind of incomprehensions have already restricted by `Facades` like `BollingerBandFacade`
@agalbenus the develop branch will alway be behind the master because the head of the master branch points to the current release and the develop branch points to the current...
Hi @jamestomk, I do not get your point. All indicators should not change their calculated value once it has been calculated for any bar. Most indicators are extending the `CachedIndicator`...
I cannot compile your example with the current release (`master `branch) nor with the current SNAPSHOT version (`develop` branch). Can you please try to reproduce your issue with the current...
In your file you are still importing classes like `org.ta4j.core.TimeSeries`. Those classes have been renamed in `develop` as well as in `master` branch. That means you are not using the...
Hi @jamestomk, thanks for your example. I think this is a bug related to the caching mechanism of `RecursivedCachedIndicator` that is extended by `ParabolicSarIndicator`. I think your example can also...
Interesting approach with some benefits. But we need to change a lot of code
hi @Andre0711er thanks for reporting, I will have a closer look
@MarcDahlem I am happy to see that there is a possible fix. I remember that there were discussions about this indicator years ago.. We should definitely fix it in ta4j...
The current implementation is not wrong nor is there a bug. You are talking about the question if the comparison is _accurate enough_. Since start of `DoubleNum` we are using...