Forward Testing Discussion Thread | Page 4 | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

MT forward testing and going live 5 replies

forward testing flippy 42 replies

Question on forward testing 5 replies

NEW " SET & FORGET" Forward Testing for VP Day Trading System 16 replies

Forward Testing Results Only 33 replies

Attachments: Forward Testing Discussion Thread
Exit Attachments

Forward Testing Discussion Thread

  • #61
  • Jan 31, 2006 10:21pm Jan 31, 2006 10:21pm
Yeah it takes a while, but the process should become faster with practice and familiarity.

Hey Rocket........how about an excel sheet set up so we can just enter the p/h and p/l that will calculate the entry price, s/l and t/p for each startegy??? Not asking too much, am I???

On another note..........sorry drjoe to hear that tragedy has struck close to home!!! You have our sympathies and support.


Quoting drjoe4x
WOWWWW!!! That took some time but all orders entered (24 here too)
Any quicker way to do that...? If not, will just hang in there

I will try to stay on top of this but am dealing with a rather close tragedy since yesterday...Remember, be thankful for everything...we don't know when our time is up, so be thankful and your happiness will follow...

good luck to all

drjoe
  • #62
  • Jan 31, 2006 10:22pm Jan 31, 2006 10:22pm
thanks again for your help!

Oh by the way....see above post


Quoting Rocket
Here are the files with date format changed to number format for Ni/pH/pL
  • #63
  • Feb 1, 2006 3:36pm Feb 1, 2006 3:36pm
Quoting BeachBum
Hey Rocket........how about an excel sheet set up so we can just enter the p/h and p/l that will calculate the entry price, s/l and t/p for each startegy??? Not asking too much, am I???
Possible that we need to look at what we want to collect as well ... the % results may not be as relevant as how many trades opened closed in profit ect..

Here is a worksheet that may help for now .. you will have to make changes to suit your strategy.
Attached File(s)
File Type: doc pHpLmedium.doc   28 KB | 608 downloads
  • #64
  • Feb 1, 2006 4:28pm Feb 1, 2006 4:28pm
Strategy #5 for 'set & forget' VP trades ended the day with a minus $801.49.

However....... IF (and that's always a big word) I had managed my 12 trades today, I would have ended with a profit. Several times during the day when I checked the trade status I was well into profit, but either had the trade reverse before the set t/p or had to close with a loss at 5pm after having been in profit earlier in the day.

My point this early on is: VP data is very useful when and if (there's that word again) you can manage your trades during the day.

But since we are looking for a good 'set & forget' system using VP.........we shall press on. One day does not make or break a system. Almost time to download and start day two...............
  • #65
  • Feb 1, 2006 6:19pm Feb 1, 2006 6:19pm
Strat #1
-226 pipsat end of day ... at around 6:30am this startergy was +100
  • #66
  • Feb 1, 2006 7:38pm Feb 1, 2006 7:38pm
I've posted the breakdown by currency over in the results only discussion. Do we want to post there daily? Or weekly?
  • #67
  • Feb 1, 2006 8:24pm Feb 1, 2006 8:24pm
Quoting spudbox
I've posted the breakdown by currency over in the results only discussion. Do we want to post there daily? Or weekly?
I noticed that you had a short and long .... were we not supposed to cancel the second trade if one got hit?
  • #68
  • Feb 1, 2006 9:54pm Feb 1, 2006 9:54pm
yes, when one trade is entered, cancel the opposite trade



Quoting Rocket
I noticed that you had a short and long .... were we not supposed to cancel the second trade if one got hit?
  • #69
  • Feb 1, 2006 9:56pm Feb 1, 2006 9:56pm
not everyone may be able to post daily..............if you want to, OK
otherwise a weekly post of all the results would be more orderly.



Quoting spudbox
I've posted the breakdown by currency over in the results only discussion. Do we want to post there daily? Or weekly?
  • #70
  • Feb 2, 2006 6:59am Feb 2, 2006 6:59am
Quote
I noticed that you had a short and long .... were we not supposed to cancel the second trade if one got hit?
oops. I'll start doing that tonight. I will need to set up access at the day job and don't have the demo account info at work.
  • #71
  • Feb 2, 2006 8:23am Feb 2, 2006 8:23am
Quoting BeachBum
not everyone may be able to post daily..............if you want to, OK
otherwise a weekly post of all the results would be more orderly.
I am concerned that if we have posts going in daily we will have information overload and it will become a jumble of data.

I think we need to decide what we want to see and how to sort it into a weeky report.

And what is the goal? I would say it is to find the best stratergy and the best pairs to trade that strategy.

So on a weekly basis we need to know how many trades per pair and weather they where in or out of money ..

please help me here because I think how we accumulate and analyze the data is the key.

Also for a weekly spreadsheet what info do we want to see. Who will do the work to log the data .. just placing the 24 trades is a part time job so we need to be able to manage the workload as well.

I think we should see some good success ... happy tradeing
  • #72
  • Feb 2, 2006 8:42am Feb 2, 2006 8:42am
Looked at charts, USD/CAD sell had missed by 1 PIP a 40+ pip gain earlier this morning...mechanical, mechanical, mechanical...I think a end of week report would be fine for each pair with each strategy...key points to me are

1. Total Trades Short/Long
2. Total Pips Short/Long
3. Trend Definition

To me, those are the most important for forward testing a completely mechanical system. Hopefully, one of the other set-ups grabbed those 40+ pips today.

good luck

drjoe
  • #73
  • Feb 2, 2006 9:17am Feb 2, 2006 9:17am
Quoting Rocket
.. just placing the 24 trades is a part time job so we need to be able to manage the workload as well.

I think we should see some good success ... happy tradeing
I was able to make some progress in the time by simply by modifying the unopened trades and placing 8 new limit orders...still took some time, but seemed to go a little quicker...just don't cancel orders, just modify if you aren't already doing so.

drjoe
  • #74
  • Feb 2, 2006 9:26am Feb 2, 2006 9:26am
I agree that we need a solid weekly report.

Rocket, if you put together an excel workbook that includes the following from each strategy as suggested by drjoe:

1. Total Trades Short/Long
2. Total Pips Short/Long
3. Trend Definition

I will enter the data if it can be emailed to me by Saturday evening at the latest. This way everyone can finish their week's data processing on that day and I have Sunday to put together the weekly report and get it posted.

The simpler the weekly report the better:

Strategy used, trades, pips (gained or lost) and trend seem to be what we are looking for here.
  • #75
  • Feb 2, 2006 9:35am Feb 2, 2006 9:35am
My MT terminal is acting up. Once yesterday and once today I have had two trades positioned to execute at the same price level. When that price is reached, one trade is executed and the second identical trade is not, it goes directly into my account history as if the trade has been cancelled prior to entry. This is frustrating because I will have to manually do the calculations for this 'phantom' trade in order to have my trading log complete.

Anyone else experience this problem?
  • #76
  • Feb 2, 2006 9:37am Feb 2, 2006 9:37am
Damn............it just happened again. That's two trades that did not execute today on schedule!!!!!!!!!!!



Quoting BeachBum
My MT terminal is acting up. Once yesterday and once today I have had two trades positioned to execute at the same price level. When that price is reached, one trade is executed and the second identical trade is not, it goes directly into my account history as if the trade has been cancelled prior to entry. This is frustrating because I will have to manually do the calculations for this 'phantom' trade in order to have my trading log complete.

Anyone else experience this problem?
  • #77
  • Feb 2, 2006 12:26pm Feb 2, 2006 12:26pm
I have attached a simple spreadsheet I am using to enter the VP daily data that calculates my trade entry prices for Strat #5.

The formulas can easily be modified to be used with the other strategies. This has saved me a lot of time with order entries (cut the time in half!).

Hope this helps someone...............
Attached File(s)
File Type: zip VPdataform.zip   3 KB | 526 downloads
  • #78
  • Feb 2, 2006 7:45pm Feb 2, 2006 7:45pm
Here is a file we can use for weeky results

I see this being filled out once per week, and saved by date. So that we have 4 per month.

open to suggestions..
Attached File(s)
File Type: zip weekly results.zip   4 KB | 507 downloads
  • #79
  • Feb 2, 2006 8:51pm Feb 2, 2006 8:51pm
Quote
QUOTE=Rocket]I am concerned that if we have posts going in daily we will have information overload and it will become a jumble of data.
Not at all. If the daily results are recorded for each day you only have one trade per day for each currency for each strategy. That is hardly a jumble. If we don't record on a daily basis we eliminate patterns that may emerge based on day of week. There are many trading systems that take advantage of things that occur more frequently on specific days of the week than others. Once the stats that give us that daily data are eliminated we cannot get them back without doing another test. Finding price patterns is work. Let't not throw out important information that can only be recreated by going back and doing it again. We are only talking about 20 trades for each currency for each person to note down.

Quote
I think we need to decide what we want to see and how to sort it into a weeky report.
That is great to summarize but let's not eliminate the daily information. You have to note it down anyway.

Quote
And what is the goal? I would say it is to find the best stratergy and the best pairs to trade that strategy.
The goal is to follow the rules and analyse the results after the forward testing is done. Let's not begin any analysis until the testing is completed. Viewing the data through various lenses at this stage is sure to lead to premature changes which is what we don't want.

Quote
So on a weekly basis we need to know how many trades per pair and weather they where in or out of money ..
I don't think that is the only thing we want to know. I think we want to have access to the data that results from the testing and first discover whether there is profit from any of the strategies.

Quote
please help me here because I think how we accumulate and analyze the data is the key.
Right on the money. If all you do is summarize you have much less data to work with. If you have all the data plus you summarize then you have something additional to work with.

Quote
Also for a weekly spreadsheet what info do we want to see. Who will do the work to log the data .. just placing the 24 trades is a part time job so we need to be able to manage the workload as well.
Nobody said it was easy. It does get faster and easier entering the trades after about a week of doing it.

Quote
I think we should see some good success ... happy tradeing
The only way you can fail at this experiment is not to follow through with it. The goal was never to make a profit in this exercise but instead to see what happens. Analysis comes later. That is why it is important to have daily data. Summaries tell very little. It may be that daily data tells too much but we can summarize to filter that out.

FT
  • #80
  • Feb 2, 2006 10:20pm Feb 2, 2006 10:20pm
Quoting fijitrader
Not at all. If the daily results are recorded for each day you only have one trade per day for each currency for each strategy. That is hardly a jumble. If we don't record on a daily basis we eliminate patterns that may emerge based on day of week. There are many trading systems that take advantage of things that occur more frequently on specific days of the week than others. Once the stats that give us that daily data are eliminated we cannot get them back without doing another test. Finding price patterns is work. Let't not throw out important information that can only be recreated by going back and doing it again. We are only talking about 20 trades for each currency for each person to note down.

That is great to summarize but let's not eliminate the daily information. You have to note it down anyway.

The goal is to follow the rules and analyse the results after the forward testing is done. Let's not begin any analysis until the testing is completed. Viewing the data through various lenses at this stage is sure to lead to premature changes which is what we don't want.

I don't think that is the only thing we want to know. I think we want to have access to the data that results from the testing and first discover whether there is profit from any of the strategies.

Right on the money. If all you do is summarize you have much less data to work with. If you have all the data plus you summarize then you have something additional to work with.

Nobody said it was easy. It does get faster and easier entering the trades after about a week of doing it.

The only way you can fail at this experiment is not to follow through with it. The goal was never to make a profit in this exercise but instead to see what happens. Analysis comes later. That is why it is important to have daily data. Summaries tell very little. It may be that daily data tells too much but we can summarize to filter that out.

FT
I'm a novice at this type of testing and because five of us are spending a lot of time and putting forth a lot of effort, I don't want to waste anything we gather that would be pertinent to our test, the results and the conclusions reached.

Therefore, I want us to follow FT's lead as to how we should assemble the data and analyze it. I don't want to have to do this same test again, but rather move forward based on the results of this first test period.

I truly believe we will learn a lot that will be very useful to us as traders and hopefully for those who are watching (I have received PMs telling me that others are watching what we are doing from the sidelines).

FT.......we will look to you for final direction about assembling and analyzing the data gathered.

And I want to extent my gratitude to Rocket, robrob, drjoe4x and spudbox for their efforts with this project! You guys ROCK!! The best of trading to all of you!!!
1 trader viewing now