HGI used with Hurst charts

[pacinvest]

Pedro I would really appreciate it if you would share a bit a background on how you came to use the 5.11 TMA centered lines and what you have learned about their derivation and use. As TMA calculations lie at the heart of BOB's methods and HGI I would like to know more how you got to this point.

And Thanks Pedro I usually these days ignore 'new' indys but this one has one a place on my charts this week. So valuable with HGI Thanks

[Pedro2]

pacinvest wrote:Pedro I would really appreciate it if you would share a bit a background on how you came to use the 5.11 TMA centered lines and what you have learned about their derivation and use. As TMA calculations lie at the heart of BOB's methods and HGI I would like to know more how you got to this point.

And Thanks Pedro I usually these days ignore 'new' indys but this one has one a place on my charts this week. So valuable with HGI Thanks



Okay on why I picked the 5/11 thing. Rather simple really. I am following Hurst and his research that he did in the 1970's. Hurst never referred to TMA's, but to Displaced moving averages with extensions. From my knowledge the TMA had not been used at that time.

What Hurst did was use a SMA of say 20 periods and displace it half that 10 periods back. Then he would get a second SMA. Double that of the first. 40 periods and displace that 20 periods back. Half of 40. Then he would extend the ends of the two MA lines now on the chart. Of course they crossed into the future on the chart. He found that the extended lines where they crossed was always half way up or down a trend run. It certainly was a unique discovery.

So being the chap that he was he wrote a book about it of which most people could not understand. Because being the Mathematician that he was it became two complex for the average person in high powered maths. And traders did not have computers in those days much anyway. So his book ended up collecting dust on a shelf until Peter Amaral of Trading Fives who does some programing a few years ago read about Hurst and wrote a (visual basic?) program I think to chart his idea. He sold this software to chart Hurst and a couple of other ideas using CSV data. Brilliant stuff. I actually bought the program and used it for sometime after I learnt how to save Empty4 data in CSV easily. By the way it still works great.

Now it was fairly manual thing doing this with Peter's software and wished there was a Empty4 indicator to do all this. Not too long ago I found a Hurst indicator written by a French person creating bands with a center line. It is the indicator you are using now. But the center line did not extend. But looking at the code found it to be a TMA indicator. So I used TMAcentered to extend that line and also found that posting only on open and not close kept repainting to a minimum. Now at this stage the Hurst indi was optimized at 5.

But the truly big discovery was to come. I accidentally put a second TMAcentered line of 10 periods on the chart. Which is what is nearly what is used now. I just casually looked at the Empty4 chart and noticed that. Heck that looks like Hurst!!!!! So I got the Hurst software written by Peter A and did a few days comparisons of Peter's Hurst software and Empty4. I can tell you now I was blown away as I realized I had found how to create Hurst DMA's very simply on Empty4. Hurst DMA's are really TMA's on Empty4. So for me it was one heck of a discovery. Now Hurst 2 ma's are normally double. So if the first TMA is 5 the second should be 10 periods. But I found by trial and error that adding one extra period to the 10 TMA the crosses seemed more accurate to me. The 5 was chosen first because it gave a fast smooth TMA. Therefore the second should be 10 periods. But 11 is a whisker better in my opinion.

So now you know how I created Hurst with Empty4 TMA's. But I kept it all to myself as I thought no one would be interested in my little ideas. Viewing Steve's forum for sometime and not posting and keeping to myself did make me feel a little guilty at times. Then I saw Bob's HGI. I liked it and was impressed, but saw a few members really liking it but struggling with it. I did to for a short while until I put my Hurst idea on it. I was rapped then I can tell you as it made HGI so much more powerful in my eyes. After that I wondered if I should post this idea. After a bit of a yes/no, struggle as I do not post, I decided to contribute the idea come what may even if I was laughed at. Maybe someone would think like me? I originally posted it on the HGI thread. But Macman thought it better on another thread and set the new thread up for. It was a very good idea. That is all there is to it Vince. Hope I have not bored you to much or other members with this post. The history of how I created Hurst DMA's extended on Empty4.

Pedro

[pakeha]

A great discovery Pedro. Congratulations.

I have also played around with Hurst, but definitely did not persevere with it like you have. I have traded the HGI/HURST combo on H4 over the past week or so and am starting to get the hang of it - in fact now feeling confident on this timeframe. I am not as successful on H1 at the moment, so need to work a bit more with it. Haven't tried lower timeframes at all.

Cheers

[SpiderX]

Hi Pedro,

this might be covered somewhere else in this discussion...but would like to confirm..
so based on your experience..TMA and Hurst are almost the same ?
i.e it does not make a difference which is being used ?

having been working on the EA for this, since Steve mentioned in PM that he would not be working on it.
Currently still working on the bugs before releasing it.

Cheers

[nicolkoekoek]

..."
So now you know how I created Hurst with Empty4 TMA's. But I kept it all to myself as I thought no one would be interested in my little ideas. Viewing Steve's forum for sometime and not posting and keeping to myself did make me feel a little guilty at times. Then I saw Bob's HGI. I liked it and was impressed, but saw a few members really liking it but struggling with it. I did to for a short while until I put my Hurst idea on it. I was rapped then I can tell you as it made HGI so much more powerful in my eyes. After that I wondered if I should post this idea. After a bit of a yes/no, struggle as I do not post, I decided to contribute the idea come what may even if I was laughed at. Maybe someone would think like me? I originally posted it on the HGI thread. But Macman thought it better on another thread and set the new thread up for. It was a very good idea. That is all there is to it Vince. Hope I have not bored you to much or other members with this post. The history of how I created Hurst DMA's extended on Empty4.
Pedro

Well THANKS Pedro! I am surely happy that you posted your idea, since I really like it and see the big advantage of having it! Your generosity of sharing is highly appreciated!

Nicolaas

[Nidolap]

Hi Pedro and folks

I am fascinating with testing Hurst lines and HGI and have HGI demo going on upper time frames, but what can this system do on lower timeframes?

First after trading and running demos on M5 and M15 I found some times when the indicator is working well BUT most of the time the indicators are wallowing around and are EA killers giving late entries early exits etc. This is pronounced when PA slows down.

BUT does HGI fail in all conditions? Is there a market condition where it works well on the lower timeframes?

This morning NZ time I was manually trading the FOMC announcement on the EU on M5 or trying to. Big moves big candles and potential disasters. I then tried looking at my HGI - Hurst lines template on M1 and to my surprise it was handling the high volatility extremely well. In fact better than any timeframe in terms of being able to see entries and exits. A possible explanation here is that within those news spikes and whipsaws is in fact short intense waves that the HGI and Hurst system can detect efficiently. Why not? And is HGI and Hurst lines clever enough to detect these waves? I think so if we know where to look.

OK have a look at the screen shot and then try looking at this system on the 1m chart sometime when the volatility gets cranking. I am going to have a go manually trading this at one of these newsy times.

Could we turn Steve's HGB EA loose on M1 during a news period?

Don't mortgage your wife on this trade concept but do have fun I am..........................

Seems very interesting as a scalping method, I eyeballed some days this week on eurusd and saw that the hurst really separates the good from bad signals. Will have a look in it, thanks for sharing this idea!

[BobbyT]

Hi Pedro,

this might be covered somewhere else in this discussion...but would like to confirm..
so based on your experience..TMA and Hurst are almost the same ?
i.e it does not make a difference which is being used ?

having been working on the EA for this, since Steve mentioned in PM that he would not be working on it.
Currently still working on the bugs before releasing it.

Cheers

You've made my day. Cheers big ears

[Pedro2]

Hi Pedro,

this might be covered somewhere else in this discussion...but would like to confirm..
so based on your experience..TMA and Hurst are almost the same ?
i.e it does not make a difference which is being used ?

having been working on the EA for this, since Steve mentioned in PM that he would not be working on it.
Currently still working on the bugs before releasing it.

Cheers

SpiderX,

You asked about whether TMA (centered) and Hurst are the same. Good question. I say yes according to my tests with software that plotted DMA's exactly like Hurst did, and TMAcentered MA's plotted on Empty4. Because I had this software that plotted DMA's like Hurst did and Empty4. I could run both at the same time and I could not really tell any difference between the Displaced Moving Averages and Triangular Moving Averages. Remember that I could only get snap shots off the Hurst software where as Empty4 I had real time operation. That real time I truly liked. Hurst really mainly concentrated on daily charts. 240m works well to down to 60m. But one needs more caution if going below that in my opinion. So hope that helps?

Pedro

[dclayw]

Have been testing HGI with the Hurst add-on and looks promising. Thanks for posting Pedro, I think it's a very valuable contribution you have made.

Here is a chart on a recent trade on the 4hr on cadchf. I don't want to get into the re-painting arguement but I just wanted to get an idea on how you would handle the exit on such a trade.

Entry was on the big red arrow near the close of the bar (around 0.8500 as shown on the chart). I did not set a TP based on TMA cross as the TMAs were near parallel, HGI was also posting red RADs all the way down. This trade got to 300 pips in profit. Plan was to exit on a yellow wave. Well the yellow wave did come, I've shown on the chart with a cross where it popped up, that yellow wave has now gone. Also this yellow wave did not appear on the chart 'till price was around 0.8360, some 4-5 bars later, with half that 300 pips already gone.

You can also see a big green arrow/blue wave combo which shows the start of a nice up move. Those symbols did not appear on my chart 'till the following day, after Empty4 shutdown/restart.

In hindsight one can come up with many better areas to exit, but I'm looking at this purely from the HGI/Hurst perspective, so I was letting it run, hoping for the yellow wave. Pedro (and others) how would you handle an exit in this situation. Yellow wave came much too late and much of the profit has gone, green arrow/blue wave combo came much later again. Would you exit when the faster moving solid TMA started turning up?

I have 3-4 other examples of similar type trades where the exit symbol (yellow wave) comes much too late and eats up too much of the open profit. The posting of the green arrow/blue wave combo would have also given an exit and a nice entry in the opposite direction but those symbols did not appear 'till many bars after the exit bar at 0.8359. Unfortunately this entry off the green arrow/blue wave would have been missed, looks very good as history on the chart though.

[pakeha]

dclayw

Good post. I have experienced the same thing.

One option, which I use on higher timeframes, is a bband stop or a pricechannel stop indicator to signal the exit. In your case the exit would have been about 0.8290 (pricechannel stop with channel = 7), so would have saved you roughly 70 pips at best. Note: don't use these indicators for entry, but rather as a SL/exit indicator.

Just a thought, especially on a very healthy trade like the one you had.