Bob and Shelley Again

[remix919]

Just curious, is there any reason a lot of people use iHistory to view the results instead of just adding their account to myfxbook?

[fchaman]

Personally my opinion about my trade performance for the week is what is my Net P/L and Net Pips...That's all I care to see in a glance...this is why I only prefer to use iHistory!

[SteveHopwood]

Just curious, is there any reason a lot of people use iHistory to view the results instead of just adding their account to myfxbook?

Why should people link their accounts to myfxbook? Giving their live account details to a third party webpage? Hmmmmm. Wouldn't give my bank account details to a third party webpage etc

I agree with fchaman. It is the bottom line that counts and as my experiences with BASA taught me, even the pip count can be irrelevant - I have had BASA weeks with positive pips but negative cash. Trading 10.6 as I do this cannot happen; iHistory gives an accurate picture of my bottom line over the period I choose to view, and this is all I need.

[SpiderX]

Hi ,

not sure if this has been brought up before, but pretty sure this logic is broken or incomplete when using BASA on 1h charts.

Code: Select all

  if (AllowMainEntry)//5. Confusing, I know, but that is how bots develop
         if (Close[2] < FastTrendMaVal[2])//1
            if (Close[1] > FastTrendMaVal[1])//2
               if (RsiPinStatus == pinlow)//3
                  if (OpenTrades == 0)//4
                     if (!useMaStackForMainTrade || MaStackStatus == stacklong)//6
                     {   
                        MainEntry = true;
                        SendLong = true;
                        TradeComment = MainTradeComment;
                     }//if (!useMaStackForMainTrade || MaStackStatus == stacklong)

From Lookfortradingopportunities function....
In the 1H chart, this implies, the Close of 2 Candles(1H) must be less than the 60MA from 2 Candles(4H) ago.
My logical understanding is that this can be right or wrong depending on the time of your hourly candle.

For example:

60Ma values (4H)
0000-0400 1.01 [2]
0400-0800 1.02 [1]
0800-1200 1.03 [0]

On 1H, Relevant 60MA Values cascaded from 4H
0700-0800 1.01 cascaded from 0000-0400
0800-0900 1.02 cascaded from 0400-0800
0900-1000 1.02 cascaded from 0400-0800
1000-1100 1.02 cascaded from 0400-0800
1100-1200 1.02 cascaded from 0400-0800

Assuming Candle shift 0 on 1H is 0900-1000, then it is correct when we compare Close[2] (0700-0800) with 60Ma (0000-0400) (value 1.01) on the 4H.
However, when shift 0 is 1000-1100, then shouldnt we be comparing Close[2](0800-0900) with 60Ma (0400-0800) (1.02) on the 4H , since this is the relevant 60Ma on the higher TF ?

Proposed fix:

Code: Select all

         int OffSet=0;

         for (cc = 1; cc<= 3; cc++)
         {
            RsiTmaVal[cc] = GetRsiTma(cc);
            if (RsiTmaVal[cc] >= PinHighLevel) RsiPinStatus = pinhigh;
            if (RsiTmaVal[cc] <= PinLowLevel) RsiPinStatus = pinlow;

            if (cc < 3)
            {
	       if(Period()!=FastMaTdPeriod)
	       		OffSet=iBarShift(Symbol(), FastMaTdPeriod, Time[cc]);
	       else
			OffSet=cc;

               FastTrendMaVal[cc] = GetMa(Symbol(), FastMaTdTF, FastMaTdPeriod, FastMaTdShift, FastMaTdMethod, FastMaTdAppliedPrice, OffSet);
            }//if (cc < 3)            
         }//for (cc = 1; cc>= 3; cc++) 

Apologies if the explanations are not clear.
Well, my explanations could be wrong if that is the intended manner for BASA to work in the 1H timeframe.

Cheers

[fxozgirl]

Hi ,

not sure if this has been brought up before, but pretty sure this logic is broken or incomplete when using BASA on 1h charts.

Code: Select all

  if (AllowMainEntry)//5. Confusing, I know, but that is how bots develop
         if (Close[2] < FastTrendMaVal[2])//1
            if (Close[1] > FastTrendMaVal[1])//2
               if (RsiPinStatus == pinlow)//3
                  if (OpenTrades == 0)//4
                     if (!useMaStackForMainTrade || MaStackStatus == stacklong)//6
                     {   
                        MainEntry = true;
                        SendLong = true;
                        TradeComment = MainTradeComment;
                     }//if (!useMaStackForMainTrade || MaStackStatus == stacklong)

From Lookfortradingopportunities function....
In the 1H chart, this implies, the Close of 2 Candles(1H) must be less than the 60MA from 2 Candles(4H) ago.
My logical understanding is that this can be right or wrong depending on the time of your hourly candle.

For example:

60Ma values (4H)
0000-0400 1.01 [2]
0400-0800 1.02 [1]
0800-1200 1.03 [0]

On 1H, Relevant 60MA Values cascaded from 4H
0700-0800 1.01 cascaded from 0000-0400
0800-0900 1.02 cascaded from 0400-0800
0900-1000 1.02 cascaded from 0400-0800
1000-1100 1.02 cascaded from 0400-0800
1100-1200 1.02 cascaded from 0400-0800

Assuming Candle shift 0 on 1H is 0900-1000, then it is correct when we compare Close[2] (0700-0800) with 60Ma (0000-0400) (value 1.01) on the 4H.
However, when shift 0 is 1000-1100, then shouldnt we be comparing Close[2](0800-0900) with 60Ma (0400-0800) (1.02) on the 4H , since this is the relevant 60Ma on the higher TF ?

Proposed fix:

Code: Select all

         int OffSet=0;

         for (cc = 1; cc<= 3; cc++)
         {
            RsiTmaVal[cc] = GetRsiTma(cc);
            if (RsiTmaVal[cc] >= PinHighLevel) RsiPinStatus = pinhigh;
            if (RsiTmaVal[cc] <= PinLowLevel) RsiPinStatus = pinlow;

            if (cc < 3)
            {
	       if(Period()!=FastMaTdPeriod)
	       		OffSet=iBarShift(Symbol(), FastMaTdPeriod, Time[cc]);
	       else
			OffSet=cc;

               FastTrendMaVal[cc] = GetMa(Symbol(), FastMaTdTF, FastMaTdPeriod, FastMaTdShift, FastMaTdMethod, FastMaTdAppliedPrice, OffSet);
            }//if (cc < 3)            
         }//for (cc = 1; cc>= 3; cc++) 

Apologies if the explanations are not clear.
Well, my explanations could be wrong if that is the intended manner for BASA to work in the 1H timeframe.

Cheers

SpiderX...I don't really understand the coding that well, but I'm quite sure all the coding within BASA was done for the 4H timeframe which is what I designed it for. Our fellow traders have simply taken BASA in it's present form and plonked it on 1H charts and traded it as is.

[SpiderX]

SpiderX...I don't really understand the coding that well, but I'm quite sure all the coding within BASA was done for the 4H timeframe which is what I designed it for. Our fellow traders have simply taken BASA in it's present form and plonked it on 1H charts and traded it as is.

Hi,

Yes. The code is correct for 4H, but may or may not be correct depending on the behavior we desire for BASA 1H.
On the 1H chart, the 4H 60Ma, and 240 MA would not be so continuous, and would look something like the Daily Pivot in Bob's 10.6 Chart,
I believe we should take that into account if using BASA on 1H.

Cheers

[SpiderX]

Just curious, is there any reason a lot of people use iHistory to view the results instead of just adding their account to myfxbook?

Why should people link their accounts to myfxbook? Giving their live account details to a third party webpage? Hmmmmm. Wouldn't give my bank account details to a third party webpage etc

I agree with fchaman. It is the bottom line that counts and as my experiences with BASA taught me, even the pip count can be irrelevant - I have had BASA weeks with positive pips but negative cash. Trading 10.6 as I do this cannot happen; iHistory gives an accurate picture of my bottom line over the period I choose to view, and this is all I need.

Hi Steve,

..would disagree with this statement, myfxbook provides some powerful features for us to improve profitability.
There are members here who have done analysis on trade opening hours vs profitability using myfxbook to tell what trading hours to avoid in order to boost profitability.

Another example of a powerful feature as you see in the attached picture. MAE(Most adverse excursion) vs trade outcome: this was the analysis of Philipe's Marilyn.
Based on the chart, we could recommend to Philipe to set a stop loss of around 220 pips for Marilyn since most of his profitable trades dont go beyond this level in terms of maximum risk.
Yes, he might lose some good trades this way, but he would avoid more bad trades potentially losing more than 220 pips.
If someone has a link, we could be able to do similar analysis and recommendations on BASA.

Also, to the best of my knowledge, myfxbook is like showing your account statement to the world, but not exactly giving them access to your account.

Cheers

[SpiderX]

Hi,

Uploading here an edited file with the above mentioned changes for 1H timeframe.
Have edited this relatively fast without much testing, so there may be some bugs.
(but highly unlikely given the simple nature of the edit)

Cheers

Edit

20/12/13
v2q mtf:
-BASA for MTF. This makes BASA more suitable for use lower time frames: 1H, 30 mins (?), 15 mins(?)
-This operates the same as original BASA in 4H timeframe.

v2r1 mtf:
-Added variable ClAtrShift. Rightly this value should be set as 2. To compare the previous candle length to the previous 2 candle Atr. Defaults to 0, so the logic would be same as original BASA

v2r2 mtf:
-Added useLowerTf variable under General Inputs. This is false by default which disables the earlier changes made for lower time frame.
Enable true to use the modifications made for lower time frame.

23/12/13
-Default ClAtrShift to 2 as confirmed by Shelley.

-Default SpotdespotCandleCloseTf to 0
-Corrected a bug in the first edit

27/12/13

v2s mtf
-Added WapenCloseCandles, defaults to 0
-Only active when value is > 0 and overwrites SpotDespot closure, uses timeframe of SpotdespotCandleCloseTf
-Close the trade after (WapenCloseCandles) candles if price does not go above the trade open price

04/01/2014

v2s2 mtf
-Edited possible bug when WapenCloseCandles is set to 0

11/02/2014

v2t
- rebuild for Empty4 build v600
- added in StringSubStringOld function as suggest in http://www.forexfactory.com/showthread. ... ost7268388
This replaces the old StringSubString function since it is mentioned to be quirky.
-Some simple code edits to make compatible to new build.

-Added updated indicator which compiles fine on the new MQ4 editor. Changed indicator_color to indicator_color1

[Wapen]

Hi,

Uploading here an edited file with the above mentioned changes for 1H timeframe.
Have edited this relatively fast without much testing, so there may be some bugs.
(but highly unlikely given the simple nature of the edit)

Cheers

Hi, thanks. will start this on a new demo beginning of jan. The current one with the so called error are doing well on the 1h so I won't just drop it. thus will see how the edited one is doing. Thanks SpiderX.

Wapen

[trader689]

many thanks for your mod SpiderX. like Wapen, my vanilla BASA has been doing well on the H1 charts as well, so will also set up a separate demo for this in January

trader

Hi,

Uploading here an edited file with the above mentioned changes for 1H timeframe.
Have edited this relatively fast without much testing, so there may be some bugs.
(but highly unlikely given the simple nature of the edit)

Cheers