You are not logged in or registered. Please login or register to use the full functionality of this board...


Post Reply 
 
Thread Rating:
  • 1 Votes - 1 Average
  • 1
  • 2
  • 3
  • 4
  • 5
A new approach to Blessing
03-15-2011, 05:35 AM (This post was last modified: 04-07-2011 08:27 AM by isaacp.)
Post: #1
A new approach to Blessing
I want to suggest a new approach for trading blessing and for set files configurations.
Instead of looking for the most secure settings for blessing, looking for setting that will never hit equity protection. (based on the past, nothing is safe for the future).
We can look for the best settings by profitability (100% for the period - a month or even week), when the chance to get to account blow up is up to 20% (period one to five on average).
The statistical expectancy of this set file is ( %equity growth on profit * profit chance ) + ( %equity growth on loss * loss chance) = (2 * 0.8) + (0 * 0.2) = 1.6 = 60% profit per month.
BUT, we should remember that after an account blow up we do not have money for trading.
So, we need to diversify our account to trade 8 - 10 pairs, and even more that 1 set file per pair ( at the same TF or at other TF) and especially for set file that has a low correlation.
then we can expect a profit of 60% per period.

(add at 3/17 6:13 GMT)
I think that I need to emphasize something important.
The sets does not have to be steady,but the average DD for its profit has to be low.
For example, if a Set showing a profit of 800 and Max DD is 1000 so you need at least 2000$, but for this method if the same set has an average DD of 200, and the big DD is only once or twice, its Enough 800$. so the profit will be 100% Instead of 40%.
Of course the risk goes up, but that we eliminate by diversification.
Each set should work on a separate part of the money, and only at the end of each period, the money is Re-balanced again.

How to check an average DD? (The MT4 can not do it)

(add at 3/22 22:25 GMT)
I develop an library that to it, look at post 24 for Details.
Run it With MM = false and start lot = 0.01 with 1000000$, after the test you will see the "nornal DD"

(add at 4/4 12:25 GMT)
on post 25 I add a new version of the dd indicator

(Add at 7/4 11:25 GMT)
om post 26 I add a new version of the indicators, now it works also on live data


Attached File(s)
.mq4  blessing_3_v3.9.2 WithGV.mq4 (Size: 103.97 KB / Downloads: 58)
Find all posts by this user
Quote this message in a reply
03-15-2011, 05:44 AM
Post: #2
RE: A new approach to Blessing
(03-15-2011 05:35 AM)isaacp Wrote: You are not allowed to view links. Register or Login to view.So, we need to diversify our account to trade 8 - 10 pairs, and even more that 1 set file per pair ( at the same TF or at other TF) and especially for set file that has a low correlation.

Correlation is on the pair, not setfile. Blush

Look at japan - the Yen is tumbling, so EURJPY, USDJPY and GBPJPY are on the downtrend. If you have trades on these three, loss will be multiplied by 3 and your account may go bust 3 times faster.

So you must find pairs that do not correlate and you will only have 3 to 4 pairs.

The Titanic was built by professionals, but Noah's Ark was built by amateurs.
The staff does not supply assistance via private mail.
Please open a new discussion in the appropriate section and you will be answered.
Find all posts by this user
Quote this message in a reply
03-15-2011, 09:55 AM
Post: #3
RE: A new approach to Blessing
(03-15-2011 05:44 AM)The Tramp Wrote: You are not allowed to view links. Register or Login to view.
(03-15-2011 05:35 AM)isaacp Wrote: You are not allowed to view links. Register or Login to view.So, we need to diversify our account to trade 8 - 10 pairs, and even more that 1 set file per pair ( at the same TF or at other TF) and especially for set file that has a low correlation.

Correlation is on the pair, not setfile. Blush

Look at japan - the Yen is tumbling, so EURJPY, USDJPY and GBPJPY are on the downtrend. If you have trades on these three, loss will be multiplied by 3 and your account may go bust 3 times faster.

So you must find pairs that do not correlate and you will only have 3 to 4 pairs.

In this situation You are right, but in normal situation You can find a set files that are not sensitive to the same move of price - if You change the enter signal indicator, or if You use an separate sets for long and short.
Find all posts by this user
Quote this message in a reply
03-15-2011, 10:26 AM
Post: #4
RE: A new approach to Blessing
(03-15-2011 09:55 AM)isaacp Wrote: You are not allowed to view links. Register or Login to view.
(03-15-2011 05:44 AM)The Tramp Wrote: You are not allowed to view links. Register or Login to view.
(03-15-2011 05:35 AM)isaacp Wrote: You are not allowed to view links. Register or Login to view.So, we need to diversify our account to trade 8 - 10 pairs, and even more that 1 set file per pair ( at the same TF or at other TF) and especially for set file that has a low correlation.

Correlation is on the pair, not setfile. Blush

Look at japan - the Yen is tumbling, so EURJPY, USDJPY and GBPJPY are on the downtrend. If you have trades on these three, loss will be multiplied by 3 and your account may go bust 3 times faster.

So you must find pairs that do not correlate and you will only have 3 to 4 pairs.

In this situation You are right, but in normal situation You can find a set files that are not sensitive to the same move of price - if You change the enter signal indicator, or if You use an separate sets for long and short.

Non-correlated pairs. I seem to remember someone from this forum did an extensive study on this subject some time ago. Can anyone access the results of that study? As of right now I trade, GBP/USD, Eur/Yen and Aud/Nzd. That seems to take care of all the majors with the exception of the CHF. Any suggestions would be appreciated.
Find all posts by this user
Quote this message in a reply
03-15-2011, 01:43 PM
Post: #5
RE: A new approach to Blessing
(03-15-2011 10:26 AM)FreeFall Wrote: You are not allowed to view links. Register or Login to view.
(03-15-2011 09:55 AM)isaacp Wrote: You are not allowed to view links. Register or Login to view.
(03-15-2011 05:44 AM)The Tramp Wrote: You are not allowed to view links. Register or Login to view.
(03-15-2011 05:35 AM)isaacp Wrote: You are not allowed to view links. Register or Login to view.So, we need to diversify our account to trade 8 - 10 pairs, and even more that 1 set file per pair ( at the same TF or at other TF) and especially for set file that has a low correlation.

Correlation is on the pair, not setfile. Blush

Look at japan - the Yen is tumbling, so EURJPY, USDJPY and GBPJPY are on the downtrend. If you have trades on these three, loss will be multiplied by 3 and your account may go bust 3 times faster.

So you must find pairs that do not correlate and you will only have 3 to 4 pairs.

In this situation You are right, but in normal situation You can find a set files that are not sensitive to the same move of price - if You change the enter signal indicator, or if You use an separate sets for long and short.

Non-correlated pairs. I seem to remember someone from this forum did an extensive study on this subject some time ago. Can anyone access the results of that study? As of right now I trade, GBP/USD, Eur/Yen and Aud/Nzd. That seems to take care of all the majors with the exception of the CHF. Any suggestions would be appreciated.

This issue fascinates me, because most pairs are correlated to the US dollar, but it is true that some set files are better than others at predicting the right direction and keeping very low baskets. Because every open pair will eat part of your margin, I suggest one pair running hotter (i.e. AUDNZD at 0.3 lots for $5K) while the others contribute in lots of 0.01.

The true quest should be to find ONE PAIR with two set files that are aggressive yet pass 2007 to present or any period inside that bracket, but one sells and the other one buys. If we could find two set files such as AUDNZD that trades as the current one, but one long and the other one short, then our profitability would increase tremendously.

In short, we could have 10 tabs: two set files for each pair, each doing the opposite. So that regardless of correlation, only 5 could ever be "stuck." While the other five, although eating margin, are working to increase it.

The interesting fact is that most pairs I have tested going long will not survive long periods. This not only makes no sense, but points to perhaps an inherent problem of MT4 in its backtesting code...I'm just venturing out with this idea.

Anyone care to start a group exploring set files that only go long?

G$ after a long hibernation.
Find all posts by this user
Quote this message in a reply
03-15-2011, 09:50 PM
Post: #6
RE: A new approach to Blessing
(03-15-2011 01:43 PM)gerald325i Wrote: You are not allowed to view links. Register or Login to view.
(03-15-2011 10:26 AM)FreeFall Wrote: You are not allowed to view links. Register or Login to view.
(03-15-2011 09:55 AM)isaacp Wrote: You are not allowed to view links. Register or Login to view.
(03-15-2011 05:44 AM)The Tramp Wrote: You are not allowed to view links. Register or Login to view.
(03-15-2011 05:35 AM)isaacp Wrote: You are not allowed to view links. Register or Login to view.So, we need to diversify our account to trade 8 - 10 pairs, and even more that 1 set file per pair ( at the same TF or at other TF) and especially for set file that has a low correlation.

Correlation is on the pair, not setfile. Blush

Look at japan - the Yen is tumbling, so EURJPY, USDJPY and GBPJPY are on the downtrend. If you have trades on these three, loss will be multiplied by 3 and your account may go bust 3 times faster.

So you must find pairs that do not correlate and you will only have 3 to 4 pairs.

In this situation You are right, but in normal situation You can find a set files that are not sensitive to the same move of price - if You change the enter signal indicator, or if You use an separate sets for long and short.

Non-correlated pairs. I seem to remember someone from this forum did an extensive study on this subject some time ago. Can anyone access the results of that study? As of right now I trade, GBP/USD, Eur/Yen and Aud/Nzd. That seems to take care of all the majors with the exception of the CHF. Any suggestions would be appreciated.

This issue fascinates me, because most pairs are correlated to the US dollar, but it is true that some set files are better than others at predicting the right direction and keeping very low baskets. Because every open pair will eat part of your margin, I suggest one pair running hotter (i.e. AUDNZD at 0.3 lots for $5K) while the others contribute in lots of 0.01.

The true quest should be to find ONE PAIR with two set files that are aggressive yet pass 2007 to present or any period inside that bracket, but one sells and the other one buys. If we could find two set files such as AUDNZD that trades as the current one, but one long and the other one short, then our profitability would increase tremendously.

In short, we could have 10 tabs: two set files for each pair, each doing the opposite. So that regardless of correlation, only 5 could ever be "stuck." While the other five, although eating margin, are working to increase it.

The interesting fact is that most pairs I have tested going long will not survive long periods. This not only makes no sense, but points to perhaps an inherent problem of MT4 in its backtesting code...I'm just venturing out with this idea.

Anyone care to start a group exploring set files that only go long?

G$ after a long hibernation.

Nice post G ~
I'm a little busy at the moment, but I'd be all for this...

Ratz


You are not allowed to view links. Register or Login to view. v1.3

You are not allowed to view links. Register or Login to view.
Find all posts by this user
Quote this message in a reply
03-15-2011, 11:29 PM
Post: #7
RE: A new approach to Blessing
Thanks Ratz...I'm testing orlee's AUDCAD 1M set file, it does surprisingly well both long and short. And yes the baskets get crowded but it is a single digit DD when balance is $5K. I'll post more under that pair.
Find all posts by this user
Quote this message in a reply
03-16-2011, 07:19 AM
Post: #8
RE: A new approach to Blessing
(03-15-2011 01:43 PM)gerald325i Wrote: You are not allowed to view links. Register or Login to view.In short, we could have 10 tabs: two set files for each pair, each doing the opposite. So that regardless of correlation, only 5 could ever be "stuck." While the other five, although eating margin, are working to increase it.

This is my "trade Both ways" theory. Shy

I am applying it on two non correlated pairs but what I noticed is that it goes in waves: you can get the occation that the two set files on a pair go in the same direction (one went short while PA is bullish and eventually both sets will be short while PA is bearish) but when both baskets closes they both go in the opposite direction and the cicle starts over again.

The Titanic was built by professionals, but Noah's Ark was built by amateurs.
The staff does not supply assistance via private mail.
Please open a new discussion in the appropriate section and you will be answered.
Find all posts by this user
Quote this message in a reply
03-17-2011, 03:08 AM
Post: #9
RE: A new approach to Blessing
First of all, I think that I need to emphasize something important.
The sets does not have to be steady,but the average DD for its profit has to be low.
For example, if a Set showing a profit of 800 and Max DD is 1000 so you need at least 2000$, but for this method if the same set has an average DD of 200, and the big DD is only once or twice, its Enough 800$. so the profit will be 100% Instead of 40%.
Of course the risk goes up, but that we eliminate by diversification.
Each set should work on a separate part of the money, and only at the end of each period, the money is Re-balanced again.
How to check an average DD? (The MT4 can not do it)
To do this We need to add a few lines of code to Blessing,

At the start() function we put a 2 lines that put the AccountEquity(),AccountMargin() into a global variables.
We need also an indicator to plot the data on the test chart (I will work on that later).
Then, we run a back test with MM OFF (this is very important) in visual mode, with the indicator.
I will add this to the first post also.
Find all posts by this user
Quote this message in a reply
03-17-2011, 04:29 AM
Post: #10
RE: A new approach to Blessing
hi isaacp,
this thread is very important. Because I´m an newbie to Blessing3 to find all the fitting files is very difficult. So this one is great to have all great tested .set files for Blessing3 in one place. And best should be that all files are workin together within the basket. Great idea and effort.
Thank you for this.
Dirk
Find all posts by this user
Quote this message in a reply
Post Reply 


Possibly Related Threads...
Thread Author Replies Views Last Post
  A modified Blessing to play with jlcgarcia 284 78,816 04-20-2014 06:44 PM
Last Post: jlcgarcia
  Criteria For Blessing Commercial Set Files daCoops 3 95 04-20-2014 12:57 PM
Last Post: Ratz
Question Conservative Blessing 3, SL included andreu 6 2,203 02-14-2014 04:38 AM
Last Post: Dkmillion
  Blessing BT Reports (August 1, 2010 to August 1, 2013) Anova 2 415 09-06-2013 02:50 AM
Last Post: Anova
  My Last Shot At Blessing jlcgarcia 18 3,406 06-12-2013 01:52 AM
Last Post: leorczhu
  Blessing grid exit blaisen1 3 637 01-18-2013 01:00 AM
Last Post: Ratz
  What's your approach to handling this problem? Catabolist 3 465 12-10-2012 11:53 AM
Last Post: SteadyEddie
  How to reverse blessing sirajali 3 983 08-10-2012 11:20 PM
Last Post: Ratz
  Best blessing setting I found out . curtischang 7 4,905 07-28-2012 06:03 AM
Last Post: The Tramp
  set file for Blessing 3.9.6 krantony 5 5,194 07-26-2012 08:34 AM
Last Post: rwb181

Forum Jump:


User(s) browsing this thread: 1 Guest(s)

 Quick Theme:



Products
Set File Database, B3Pro! and more on the way

Affiliates/Legal
Terms and Conditions
Privacy Policy

Online Users
There are currently 19 online users.
» 4 Member(s) | 15 Guest(s)
aicomputerd, alasdeaguila, energyvodka

Quick Links
Home
Tools
Help Center

Search Forums

(Advanced Search)

Welcome, Guest
You have to register before you can post on our site.


  

Password