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)
Looks almost identical to mine. So I suppose like any indicator, it has to be used as a way to strengthen conviction in some idea, and not be the main reason to execute an idea. For example, it may be perfectly good to buy a new low even if A/D is making a new low as well; it need not have already put in a low and now make a higher low in order for us to feel strong about an idea. Real-time observation and use is probably the only thing that will really do the job, but tomorrow I will plan to do some visual backtesting to see if anything stands out. Trying to get to bed before 2:30am like last night, signing off for now and talk soon more on this subject!
Can you help answer these questions from other members on NexusFi?
If I set JINT.Z to 1 minute, then my sample ES chart looks like this (panel 2 is McClellan from JINT.Z, panel 4 is me testing a Moving Median of 34 on JINT.Z)
But if I switch to 30 seconds for JINT.Z, I get this:
And finally, 1 second:
Now I don't care about the absolute value of the McClellan, but I do care about the shape. A 10k volume bar on ES is much more than 1-minute, so I am trying to understand why the shape is so drastically different. Sierra Chart should be taking the value from JINT.Z that most closely matches (time wise).
Since McClellan is using a 34 period EMA, we know that a 1 second bar vs 60 second bar is going to have a much different smoothing curve. But I was not expecting the difference in results between the 60 second and 30 second as you see above. Look at the 30 second version and you'll see around 12:16PM a higher high in panel 2. But on the 60 second version, it is a lower high.
When using the Moving Median of 34, it keeps its shape regardless of time frame.
Now on the SPX.XO daily chart, I found I need a separate source chart for JINT.Z to also be set to 1 day. Sierra cannot use the tick based (second) chart as a source for the daily chart, values are all wrong.
So I have JINT.Z @ 1 day, and SPX.XO @ 1 day. Panel 2 is the McClellan Summation, and panel 3 is a moving median of 34.
Results here are as expected. So still working out in my head why the difference on the smaller chart.
The first thing I would do is add a few days on your JINT chart (3 should guarantee another trading day) to be sure that you have enough history for the lookback period of the EMA. So if days is 5, change it to 10 and see if that alters the view.
Exported using NT Version 7.0.1000.12
Following a request by a member of this forum I have quickly implemented a version of the McClellan Oscillator and the (calibrated) McClellan Summation Index. The indicators can only be used with daily data. It …
The McClellan indicator is a MACD based on net advancing/declining issues. MACDs are momentum derived indicators that already turn negative when the input series declines. The average net advancing/declining issues are still positive over the last week, but their level has already declined from earlier highs.
The McClellan oscillator is a leading indicator and turns earlier than a MACD of SPX. For that reason it can be used for such entertaining signals as the Hindenburg Omen.
Attached two charts of the McClellan oscillator and summation index.
On an intraday chart, the best approximation of the McClellan oscillator would be a MACD (13,39) of the NYSE Tick. This has been suggested by the MCClellans themselves.
The problem is that the tick data is only available with the open of the stock exchange. If you apply it to a minute chart, the absolute values are insignificant for the first hour. Actually the training period of the EMA(39) is longer than 39 bars, because the EMA is an IIR filter.
It would make sense to replace the EMA with a SMA that builds up starting with the session begin and which does not take into consideration values of the prior session of the NYSE-Tick. A modified Awesome Oscillator would be the tool of choice.
Below is a comparison of a simple EMA based intraday Tick Oscillator and a 39-period median of the NYSE-TICK. The moving median is a TPO-median, which runs smoother than a median taken from the close. The indicators use 1-min resolution data, the S&P 500 chart has 5-minute data.
