How to test the use of an econ filter in FX market; The backtest results for today's data on GBPJPY - General

brumsey 2006.03.27 22:18 #131

Just a simple question.

Does this EA work better in D1 chart than the 1hr chart?

For the D1 chart , do we use the default settings?

Thanks in advance

Basza

[Deleted] 2006.03.27 22:40 #132

I am using H4 and here are the results.

I am attaching the backtest results for today's data on GBPJPY and the actual trades. It shows that the backtest data is way off!!

Anyway, I will run with H4 another day or so and then see where we stand.

Maji

Files:

statement.htm 13 kb

strategytester_2.htm 5 kb

Data output Creating custom indicators Forced test stop: TesterStop

holyguy7 2006.03.27 23:28 #133

basza:

Does this EA work better in D1 chart than the 1hr chart?

For the D1 chart , do we use the default settings?

Basza

Good question. You be the judge. Statements on both D1 and H1 timeframes. The last 24 hours.

D1 Settings all 19 currency pairs Interbank FX offers:

SL:30

TP: 40-60

No Timefilter.

H1 Settings all 19 currency pairs Interbank FX offers:

SL:30

TP: 40-60

Longbar: 15

Timefilter only on EURUSD, GBPUSD and USDCHF

Here is something to consider, Higher Longbar appears to work better on lower timeframes. We all my want to try to test that. For example:

D1: Longbar 10

H1: Longbar 15

H4: Longbar ???

M30: Longbar ???

This is why we need to test this. I changed the longbar on the H1 to 15 and got better results. Perhaps someone could try the longbar at 17 and get even better. You see where I am going with this. Certain currency pairs react different. Perhaps longbar with timefilter might do the trick. The options are endless and that is why we need to all work together to test out many different settings.

Folks, it is the reason I posted this for FREE and why Nich has done all the work he did on the EA for FREE so that we all could test different settings and not just copy ones that I found successful. Please, freely receive, freely give.

Files:

pg_d1_sl30_all_19_pairs.htm 17 kb

pg_h1_sl30.htm 19 kb

Testing indicators Overview of Copy functions Assignment and initialization, expressions

sstillwell 2006.03.28 00:23 #134

I ran a 1-year backtest (1/1/2005 - 12/31/2005) on EUR/USD, optimizing LongBar...came up best results on LongBar=17 (optimized range 10 - 25). Starting balance = $10K, H1 timeframe. Modeling Quality = 90%

Results vary widely. Worst-case is LongBar = 25 with profit of $31460 (darn it... ) Best case is LongBar = 17 with profit of $130210 (!?!?!?! ) The profit does NOT correlate with bar size.

LongBar=17, ID_BASE=100000, lots=1, Risk=10, stoploss=30, takeprofit=40, TSactivation=40, TrailPips=5, MaxTrades=1, FromHourTrade=7, ToHourTrade=20, period=0, hour_to_close=21

I have modified the LotSize function to my own rudimentary money management...strictly chooses lot size by AccountBalance(), then iteratively scales it back down by half if there's not enough available margin. I can post it if you're interested.

This is VERY cool. Most consistent EA I've tested yet. I've set it up on a demo account for EUR/USD, GBP/USD, USD/CHF (all timeframe and weekend mode), and USD/JPY (no timeframe, weekend mode on). Let the forward testing roll on!

Scott

Strategy Testing - Algorithmic Optimization Types - Algorithmic Determining Expert Advisor parameters

laserjet 2006.03.28 01:05 #135

sstillwell:
I ran a 1-year backtest (1/1/2005 - 12/31/2005) on EUR/USD, optimizing LongBar...came up best results on LongBar=17 (optimized range 10 - 25). Starting balance = $10K, H1 timeframe. Modeling Quality = 90%

Results vary widely. Worst-case is LongBar = 25 with profit of $31460 (darn it... ) Best case is LongBar = 17 with profit of $130210 (!?!?!?! ) The profit does NOT correlate with bar size.

LongBar=17, ID_BASE=100000, lots=1, Risk=10, stoploss=30, takeprofit=40, TSactivation=40, TrailPips=5, MaxTrades=1, FromHourTrade=7, ToHourTrade=20, period=0, hour_to_close=21

I have modified the LotSize function to my own rudimentary money management...strictly chooses lot size by AccountBalance(), then iteratively scales it back down by half if there's not enough available margin. I can post it if you're interested.

This is VERY cool. Most consistent EA I've tested yet. I've set it up on a demo account for EUR/USD, GBP/USD, USD/CHF (all timeframe and weekend mode), and USD/JPY (no timeframe, weekend mode on). Let the forward testing roll on!

Scott

Hi sstillWell,

I used you setting on EURUSD, I noticed on some point there are 2 trade buy on same time like, on 2005.03.11 12:47 buy and closed with profit then there is another trade on same time buy and closed with profit.

Why are there multiple trade on same time?

Thanks in advance

Trading Principles - Trade Trading Principles - Trade Opening and Closing Positions

[Deleted] 2006.03.28 01:18 #136

laserjet:
Hi sstillWell,

I used you setting on EURUSD, I noticed on some point there are 2 trade buy on same time like, on 2005.03.11 12:47 buy and closed with profit then there is another trade on same time buy and closed with profit.

Why are there multiple trade on same time?

Thanks in advance

Laserjet,

If you study the EA very carefully then you will find that this is entirely possible. This was discussed earlier.

Suppose, at 5:15 high is 1.3000, Low is 1.2950 and open for this bar is 1.2960and ask now is 1.2961. Thus a long position is opened and the profit is taken at 1.3001. (40 pips above entry or 1.2961) However, now the high is 1.3001 and the low is still 1.2950. Now if this happens in one bar and the long entry conditions are still valid, another long entry will be fired. However, if the time of the two entries should be different, even though they are in the same bar. However, if the trade time is same, then something is wrong and a more thorough review by someone expert is needed.

Hope this helps.

Maji

Parabolic SAR - Trend Real and Generated Ticks Close By - Trade

masoudk 2006.03.28 01:45 #137

Hi holyguy7

Can anybody make an indicator base on Profit Generator Rule?

[Deleted] 2006.03.28 03:02 #138

Kohzadi,

There is are multiple EAs based on this system courtesy of holyguy and Nicholisen attached to messages earlier in this thread. Take a look at them.

Maji

holyguy7 2006.03.28 03:14 #139

I believe we all need to focus on Longbars differences and timeframes. I am finding that the longbar appears to do better with lower timeframes.

I just check the results for my original settings and my profits from yesterday are all gone today. I had a series of losses today that whipped out the profits.

However, the H1 chart with Longbar at 15 seems to still be successful. I am trying the longbar at 20 on the M15 chart as well. We need to test these settings to get some good ones.

Let's all work together to get some good settings. Longbar set to higher settings seem to make more profitable results. Let's focus on that.

Thanks,

Fractals - Bill Williams' Fractals - Bill Williams' Trading account information

holyguy7 2006.03.28 03:18 #140

kohzadi:
Hi holyguy7 Can anybody make an indicator base on Profit Generator Rule?

It probably can be done, but I would not know how to do it. It just uses price movements to execute buys and sells. I really would not know how that could be done because I am not a programmer.

Profit Generator EA - page 14