Welcome to NexusFi: the best trading community on the planet, with over 150,000 members Sign Up Now for Free
Genuine reviews from real traders, not fake reviews from stealth vendors
Quality education from leading professional traders
We are a friendly, helpful, and positive community
We do not tolerate rude behavior, trolling, or vendors advertising in posts
We are here to help, just let us know what you need
You'll need to register in order to view the content of the threads and start contributing to our community. It's free for basic access, or support us by becoming an Elite Member -- see if you qualify for a discount below.
-- Big Mike, Site Administrator
(If you already have an account, login at the top of the page)
I am putting together an indicator to send an alert when price reverts to the VWAP then, presumably, continues an existing trend.
I am finding it difficult to reduce the number of signals to a reasonable level and wondered if anyone can suggest a way to filter the conditions.
I am using a VWAP with a couple of envelopes and am looking for instances where (for a bull trend) the VWAP is rising, but a bar closes within the first envelope AND has a lower wick >= 50% of the bar's range. I also added that the low should be the lowest recent low. I still get too many signals.
I considered requiring price to be outside the envelopes a few bars ago, but that leaves out valid situations where price is staying within a narrow channel.
Any suggestions?
Can you help answer these questions from other members on NexusFi?
@DavidBodhi Your question as asked is frankly a bit odd because the first half of the question defines that you are interested in this condition and the second half argues the condition isn't very useful, after all.
The easiest way is to use a longer periodicity or send fewer alerts per day by only checking at the close of a bar. If you check only on the close of the bar then you will limit your number of alerts to the frequency of the bar.
A second and more advanced solution is to create a ranking algorithm or quality score to determine the relevance of this event. Next, you use a ranking or normalization function to rank the recent results of that algorithm and filter based on quality of signal to only send alerts for scores above certain thresholds. You would need to determine what is relevant in regards to this condition. If range markets are the problem you could create a metric that measures where price is trading in relation to recent range as a criteria.
A third option is to convert your discrete signals into continuous outputs. So, instead of having a signal that the price crosses over a VWAP instead you produce an output for every bar, similar to the MACD, and then you rank those outputs.
My wording might not have been as clear as it should have been. Certainly my intention was not to contradict myself. I was just hoping for a better 'description' of price approaching the VWAP prior to a bounce.
I'm already only checking on bar closes. I like the idea of a ranking algorhythm. I've used that technique before but it's been a while and I didn't think of it in this case. I might be able to use that to get around the fuzzinesss of my conditions.
The continuous output idea would end up being just another oscillator. I've looked at enough of those. Thanks for the reminder of the ranking concept, though.
@DavidBodhi Right, the purpose of the conversion is that it gives more signals (one per bar) and then you could run your analytics over that to understand the relationship and for thresholding purposes. There is another option which is let's say you cannot find a ranking algorithm then you could create similarity score. In this model, you determine the features you think are relevant to compute a unique score, but in this case it is not a score with any assigned value but simply measuring how much different the market is, and you only send new alerts when the similarity score is sufficiently different from the previous signal.