CoG + Ultra Trend System

what's parameters is best?m=4 is best?

Hi

I've finally find some time to implement this EA - it's attached. I've added one parameter - CoG_Modifier. if it's > 0 system opens sell orders even if price is above CoG main moving average, but not more than CoG_Modifier no of points, and accordingly for buy orders if price is below main CoG.
I haven't spend long time jet to test it but I've run quick test on GbpUsd for 2009 on default settings (which are not very accurate really) - test results attached - and it's able to earn money so looks good for me to continue work on it.
What I'd suggest which could increase income would be to add some SL trailing.
I think we could also consider to add separate params for timeframes for CoG and UT.
I also want to save some statistic informations about orders - price position against CoG may be interesting to verify CoG signals.
If you have any questions or commenst feel free to post them.

Wow this is fantastic! My expert programming skills are virtually ziltch unfortunately, hopefully I will be able to cope by the end of this year.

One thing about the current expert. Since having originally posted this system, I use the CoG more as a confirmation/barometer for other signals. Think of as a tertiary indicator really. My main signals come from the UT_Fast, confirmed by MACD, confirmed by the "Slope_Direction_Line(s)" indicator. The CoG filters the bad UT_Fast signals by throwing out movements interior to the green lines. The worthwhile, profitable signals normally occur on the greens or directly outside them.

As I said, I'm no where near as competent as you in expert programming, so it would be difficult to analyze the way the expert weights these different factors. If you could simply write an expert revolving around the Slope_Direction_Line, it would be an EXCELLENT base to add the different auxiliary and tertiary confirmation indicators.

Check this chart out of GBPUSD. If you follow the "slope direction line" (white/yellow), if becomes obvious how profitable just this type of expert alone would be. When coupled with CoG and UT_Fast filters, it would have very little drawdown.

Hi Rainingpips

Your new strategy looks really good but due to my lack of spare time at the moment I'd rather concentrate on one EA and work on it until I decide either to run it live or abandon idea.
I should find some time next week to do more testing of CoG_UT EA and then I'll let you know about results and mine conclusions.

Here's a current chart of GBPUSD from tonight that I traded from. Notice how UT_FAST began on a positive slope from a plateau below the "4" level, after there was a confirmation cross in the MACD, with bars moving to topside. Then the Slope Direction Line (35) changed to positive (white). I use a (20) and a (35) Slope Direction Line just for analysis. But with all these Indicators and confirmations, my long is good for about 52 pips ATM. Anyone noticing the doji in the middle, I set my SL at 3X ATR (20), in this case it was around 89 (I set it at 60 just to be safe) so I didn't get stopped out.

I'm going to work on programming an EA just off the Slope Direction Line indicator to build an automated system off of. Any help would be appreciated. I understand that MQL4Engineer is busy, so anyone else feel free to contribute!

Here's my GBPUSD trade from this morning, up approx. 100 pips (about 96 with the spread). UT_FAST has plateaued, so we are looking to get out of this trade, with the trend reversing. Normally, once UT_FAST has surpassed the 15.5 "median" level it will surpass the other levels and plateau above 27. It may stutter, with the trend continually moving upward, so a careful watch of the MACD is necessary to avoid getting shortchanged on the trade.

I was wondering if someone could post the UT Fast indicator.. I have been trying to get this indicator for months now and every UT i get does not work.. Can anyone help me out?

RainingPips,
It appears the Slopeline/UT-Fast combo is more cleaner and straightforward in minting pips than the COG/UT-Fast combo.


Charles

You're right, but what the CoG and MACD allows is a brief approximation of the expected magnitude of a move. For instance, today in AUDUSD.

The Slope Line (20) & (35) both went negative along the <black> "median" CoG line. It was accompanied by weak MACD activity, and consequently the move has been rather weak (my trade is break even as I type, was in the green 15 pips). Last night's move upwards had a similar location on the CoG for the Slope Line, but was accompanied by much higher MACD velocity and a steeper slope in the UT-Fast.

The main point of this post is not to discount any of these indicators as being poor performers in dictating trend direction or strength. Slope Line and Ut_fast are the main indicators to keep your eye on, but checking where they happen in the CoG and the MACD levels at the time are worthwhile confirmations.

Quick update on SL anf TP on interdaily trades.

In my AUDUSD trade, price opened above the pivot (which is normally where you want to sell), and would go to eventually bounce off the S2 (support line 2). I have found this to normally be the case (vice-versa with price hitting resistance levels), that either price action reversal or plateau occurs at these levels or midlevels, so TPs are set accordingly.

For the this trade, I set a 60% TP at S1 0.9200 (from my short @ the FPV mid-pivot of 0.9240). It broke that level, and I manually closed the rest around MS2 (midpivot btwn S1 and S2).

I'm working with a .dll import for trend following on the pairs I'm watching w/ Slope_Direction_Line. Going to try and put in an excel file.

Slope_Direction_Line Trend Pips
USDJPY: UP 45
AUDJPY: DOWN 4
GBPUSD: DOWN 40
EURUSD: DOWN 150
AUDUSD: DOWN 40
EURJPY: DOWN n/a
GBPJPY: DOWN n/a

The last two pairs haven't demonstrated sufficient UT_FAST activity to enter a trade, however they both are technically down.

One thing I've noticed is the relationship of UT_FAST to the Slope_direction_line. For instance, a slope line change NOT accompanied by a similar UT_FAST movement will probably be a false or unprofitable trade. Vice versa, a strong UT_FAST movement not accompanied by a trend change is likely to be a false breakout. The only exception that I can see to this would be a LARGE spike, that couldn't be accounted for in either indicator. This system likes consecutive momentum candles less than 40-30 pips.

Just found this thread, nice work going on!

Would you please update with your latest setup
and what is your preferred time frame now?

thanks

Brian

No problem. This scheme has changed alot since I first posted it, as it's become much less dependent on the CoG for taking trades.

Essentially, we look for confluence in the Slope_Direction_Line indicator and the UT_FAST. We see a high UT_FAST value, we are looking for shorts; and vice versa for longs. The crux of this system is the Slope Direction Line (SDL). By default, it has a period of 20, but I use two iterations; one @ (20) and one @ (35).

Trading on the 1hr period, the (35) provides a very clear picture of the daily trend, with the (20) acting as an inter-daily trend indicator. It normally takes several days for the UT_FAST to go from one extreme to another, which dictates that a trade might potentially be open for several days.

The MACD I use for confirmations. I normally look to it after I have identified a possible trend change via the SDL and UT_FAST. A cross is the best confirmation you can get, unless it gets choppy along the zero line; and in this case I would either stick with the trade held previous or have it be a no-trade.

I have integrated several features of the THV system into this method, including the "Multi-info" and the pivot system. The pivots are excellent indicators for long/short confirmations. Successful shorts are normally taken at Resistance levels, and longs at the Supports. Most of the action is going to take place between R1 and S1, but some of the higher volume pairs regularly exceed the R2 and S2 levels.

The CoG is used as a type of barometer for SDL changes. If, for example, PA has touched or exceed the top green level, and a short opportunity appears, it most likely will be a good trade. If we see a SDL change with PA hovering close to the black median level, it will be a poor performer on average.


D/L the indicators and play with it. Get a feel for anticipating trend changes (SDL changes) and then looking at the other indicators for guidance on if it will be a good trade. Ride the white and yellow SDLs for a couple of days, as on average they will go for 2-3 days before a legitimate change.

As for SLs and TPs, this is up to you. I look to the THV's "3x ATR(20)" as a good source. This is normally about 70% of the daily average (pips) and should price in any wip-saw movement. When you first take a trade, watch it manually for a few hours though. It will become very clear within 3-4 hours after opening the trade whether your interpretations of the signals was genuine.

Please make any changes you like and post your success (or failure) with the system!

One more thing on TPs for this system. I've noticed, but not thoroughly tested, that the opposite green CoG line to an SDL change acts as a good TP. For instance, on the last GBPJPY "bullish" SDL change, the top green CoG level was approx. 70 pips away from where you would have entered a long. The top of the movement was about 110, with the top hourly closed candle @ about 90. So only about 20-30 pips off the top, but better safe than sorry in my mind.

You could also through a 15,20,30 pip trailing SL in there.

Would you post the new template and any indicators fresh
in a post so we can start from there?

Much appreciated, look forward to getting into it!

Brian

Here are most of the indicators and template.

I think using the CoG with ichmoku would be a little too much on the chart. We already have the CoG and pivots for levels.

OK so a couple quickies

you like the jb cog better than your other original
cog?? yes I know that one repaints, but since
its just a support indy that should be ok.

what do you think? I think the other I can
visualize the trades better

I understand the ut fast also repaints, but
perhaps not the same issue on 1hr charts?

I have an indicator overlay and alt or
additional trend indicator to suggest
do you mind me posting a screenshot or two
for discussion?

what kind of success have you had so far


Brian

The JB CoG is better because none of the other indicators repaint, despite what you may have seen in UT_FAST. UT_FAST doesn't repaint, it just lags slightly in it's slope until the current bar closes.

So far I've had very successful trades across many currency pairs (all that were listed in another post). This system isn't a pair specific method. It relies more on volume and trend velocity to enter trades. If those two aren't present, then this system won't pump out signals. Right now I'm following the "yellow brick road" of the the downtrending Slope_Direction_Line (35) on most of the pairs. AUDUSD is good for about 70 pips if you got in right when the hourly candle closed that confirmed the downtrend.

If you have any suggestions or additions I'd be glad to see them!