Menu

Options

Bookmark Thread

First Page First Unread Last Page Last Post

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

  • #21
  • Quote
  • Edited 10:24pm Jan 25, 2006 10:17pm | Edited 10:24pm
[quote][quote=BeachBum

4 week test period
trade 6 currency pairs: (eur/usd, gbp/usd, usd/chf, usd/jpy, aud/usd, usd/cad)
enter orders between 6:00 and 6:30 pm est
close all trades the following day between 5:00 and 6:00 pm est
$ 10,000 mini demo account
one lot per trade per currency pair
trade all six pairs five days a week
don't change the s/l or the t/p during the trading day
when one order is filled for each currency, cancel the other order
if market opens outside the predicted range, do not enter that day on that currency
no money management
use Meta Trader platform (some of you may not be familiar with this trading platform. I will help anyone set this up, it's really simple to do and use and will allow all of us to be on the same trading platform. PM me for help if you want or if it's OK with FT, I'll post the help here.......
[/QUOTE]You're doing great. Thanks for all bb!

Also bb post the help you mentioned as well as the individual parameters each of the traders is going to test. I understand they'll all be different? correct?

FT
`
  • #22
  • Quote
  • Jan 25, 2006 10:26pm Jan 25, 2006 10:26pm
Quoting Rocket
Is it feasable that a standard log record is set up in excel and the participants each send the results of there assigned pair to our point of contact .. BB in this case. B could then maintain a running record and post the excell speadsheet daily or weekly ..?
I agree. If each person posts their results into exactly the same template in the same format and order it will make things very easy. Whoever is supplying the template should probably be the one to coalate the results. My 2c

FT
  • #23
  • Quote
  • Jan 26, 2006 12:29am Jan 26, 2006 12:29am
It is my understanding that we have agreed that each one of us will trade all six pairs using different settings.

For example I had selected:

enter 10 pips before P H/L
25 pip S/L
50 T/P

What are the other settings we could use for these trades? With four other participants, we need four more trading parameters with different entry points, S/L and T/P points.

Correct?
  • #24
  • Quote
  • Jan 26, 2006 5:38am Jan 26, 2006 5:38am
Quoting BeachBum
It is my understanding that we have agreed that each one of us will trade all six pairs using different settings.

For example I had selected:

enter 10 pips before P H/L
25 pip S/L
50 T/P

What are the other settings we could use for these trades? With four other participants, we need four more trading parameters with different entry points, S/L and T/P points.

Correct?
Yes.

I'd Suggest:
15 pips inside the ph/pl
35 pip stops on all and 45 on gbp
targets of 70 pips on all and 90 on gbp.

Also:
5pips inside the ph/pl
20 pip stops and 30 on gbp
targets 20 pips inside opposite s/r

Also:
10 pips inside ph/pl
25 pip stops
target of 50 pips for 1/2 lot and second 1/2 exit 25 pips inside the opposite ph/pl

What do you think bb?

FT
  • #25
  • Quote
  • Jan 26, 2006 9:19am Jan 26, 2006 9:19am
These are great FT..........thanks.

Anyone else have a trading parameter suggestion........we need one more!


Quoting fijitrader
Yes.

I'd Suggest:
15 pips inside the ph/pl
35 pip stops on all and 45 on gbp
targets of 70 pips on all and 90 on gbp.

Also:
5pips inside the ph/pl
20 pip stops and 30 on gbp
targets 20 pips inside opposite s/r

Also:
10 pips inside ph/pl
25 pip stops
target of 50 pips for 1/2 lot and second 1/2 exit 25 pips inside the opposite ph/pl

What do you think bb?

FT
  • #26
  • Quote
  • Jan 26, 2006 12:22pm Jan 26, 2006 12:22pm
Hi blue....

thanks for sharing this info, very informative!

would you care to join us for the forward test we have planned as explained recently in this thread?


Quoting bluemeasure
BeachBum & others,

One easy way to make a few pips from VP pretty much every day is by taking the half-way point btwn the Predicted high and Predicted low into account. Wherever the price starts at at 6pm EST, one can put in trades that target these values for each currency pair and have their trade come out successfully as follows (based on 3 months worth of data):

EURUSD 95.29%
USDJPY 88.10%
USDCHF 100% (!!!)
GBPUSD 91.67%

Note that if the price starts out on this mid-value as it often does, then you don't have a trade. This value does tend to get hit a number of times during a 24 hour period in ranging markets (I got those results broken down by hour somewhere as well), so when the price moves back towards this half-way value, you can enter another trade if things look right (e.g. momentum, Price Action, etc). If you and others are going through the process of forward testing, you may want to keep these half-way values in mind as it may well increase your daily pip count.

BlueMeasure
  • #27
  • Quote
  • Jan 26, 2006 12:27pm Jan 26, 2006 12:27pm
After reading the last post by bluemeasure, I suggest we set up one trade with the forward test using the T/P point being the mid-point of the daily P H/L.

Would someone like to suggest an entry point, S/L and the we will use the mid-point for T/P for this parameter?
  • #28
  • Quote
  • Edited 3:48am Jan 27, 2006 1:06am | Edited 3:48am
Quoting BeachBum
After reading the last post by bluemeasure, I suggest we set up one trade with the forward test using the T/P point being the mid-point of the daily P H/L.

Would someone like to suggest an entry point, S/L and the we will use the mid-point for T/P for this parameter?
Hi guys. I deleted some good comments from all but they were not really part of this thread topic. Had I the tools to move individual posts to another thread without it taking forever I'd just have moved them to the general discussion thread for VP.

Please don't take my deleting them as being anything other than wanting to remind people to stay on topic here.

Regardless I appreciate the importance of using VP's strengths and BM's and j4x's comments about regression trading are valid and VP is very strong on pointing out the daily mean. I left BB's copy of BM's original post for reference to some excellent stats.

Therefore to preserve the focus of this particular exercise I agree that someone should pick the entry parameters from one of the other three or four sets of parameters and simply use the mean as their target. That way we'll be capturing the results of the target from the first bounce. The time the system will suffer most is when the market is trending strong, opens around the mean and heads straight through the ph/pl.

I'd like to add one other thought that came from someone who is doing some private testing and does not have the time to participate and that observation was as follows:

When the market gaps over the weekend by say 80 pips if you add that amount or subtract that amount from the predicted daily range for the following monday it tends to reset the ph/pl surprisingly correct. Seems the daily range is predicted well but VP does not anticipate gaps. Therefore factoring in the gap is something worthy of note. I don't know if we want to include it in this test or not but I thought it worth mentioning.

FT
  • #29
  • Quote
  • Jan 27, 2006 2:09am Jan 27, 2006 2:09am
I have sent a workbook sample to BB and its possible we can begin testing 1 or more strategys next week.

Also I offered some suggestions for collecting and reviewing data .. it may be useful to throw out these suggestions to all .... as I may be off base .. I dont object at all to being steered with this .. but we do need some way to manage the information.

1. BB is the single point of contact to add strategy's, traders, improvements .. whatever needs to be done ... we need a source of leadership.
2.The workbook contains the 6 pairs suggested. It is just a basic spreadsheet for each pair. I will be happy to expand on it if the group has suggestions. This same woorkbook can be used for different strategy's .. just rename it to suit.
3. Suggestions should probably come from a single point BB .. so that he can filter and evaluate any suggestions for inclusion.
4. PM with your email and I will send a copy .. although until BB approves what I have sent. Consider it draft.
5. I have suggested to BB that we limit the number of traders to each strategy so that we dont have data overload. I am thinking that if we have 3 traders per strategy that would be enough. because (see 5.) ..
6. Each trader completes thier workbook and email a copy to myself or BB or "to be determined". If its 3 workbooks per strategy we can compare the 3 and if all are within a reasonable range of each other pick the book with the best average of the 3 for inclusion into a weekly workbook.
7. A weekly speadsheet can be prepared that summerizes results from all the tested strategies going forward. This book is sent to the participents and the comminity on request.



Im throwing this out in hopes of some positive feedback and constructive ctitique ... and get the show rolling ...[img]images/icons/icon12.gif[/img]

regards
  • #30
  • Quote
  • Jan 27, 2006 3:45am Jan 27, 2006 3:45am
Quote
QUOTE=Rocket]I have sent a workbook sample to BB and its possible we can begin testing 1 or more strategys next week.
Thanks Rocket. It's ok to post here I don't think we need to go underground with this.

Quote
Also I offered some suggestions for collecting and reviewing data .. it may be useful to throw out these suggestions to all .... as I may be off base .. I dont object at all to being steered with this .. but we do need some way to manage the information.
Agreed. I don't want people to be paranoid about contributing publicly here. It is all too easy for people to derail these projects and turn them into a perpetual development treadmill but that is probably not sufficient reason to go underground just yet.

Quote
1. BB is the single point of contact to add strategy's, traders, improvements .. whatever needs to be done ... we need a source of leadership.
Agreed. BB seems to be doing great at keeping this focussed.

Quote
2.The workbook contains the 6 pairs suggested. It is just a basic spreadsheet for each pair. I will be happy to expand on it if the group has suggestions. This same woorkbook can be used for different strategy's .. just rename it to suit.
Why not post it here so people can download and take a look?

Quote
3. Suggestions should probably come from a single point BB .. so that he can filter and evaluate any suggestions for inclusion.
Perhaps we could welcome suggestions from anyone as long as they focus within the particular project we are working on. The final call would be for bb to decide.

Quote
4. PM with your email and I will send a copy .. although until BB approves what I have sent. Consider it draft.
5. I have suggested to BB that we limit the number of traders to each strategy so that we dont have data overload. I am thinking that if we have 3 traders per strategy that would be enough. because (see 5.) ..
I figured one trader per strategy. Do we have enough participants for more than on per?

Quote
7. A weekly speadsheet can be prepared that summerizes results from all the tested strategies going forward. This book is sent to the participents and the comminity on request.
Or posted in the trading results thread.

Quote
Im throwing this out in hopes of some positive feedback and constructive ctitique ... and get the show rolling ...[img]images/icons/icon12.gif[/img]
Thanks. It seems to be working.

FT
  • #31
  • Quote
  • Jan 27, 2006 9:05am Jan 27, 2006 9:05am
ok I have attatched the file for upload.

I had thought previouse postings said everyon would trade all the pairs/strategies.
So whatever works, just need clarification.
  • #32
  • Quote
  • Jan 27, 2006 9:13am Jan 27, 2006 9:13am
ok .. here we go .. uploaded
Attached File(s)
File Type: zip VP Forward Test.zip   7 KB | 437 downloads
  • #33
  • Quote
  • Jan 27, 2006 9:27am Jan 27, 2006 9:27am
Each trader trades ALL six pairs using ONLY one strategy.

I will be consolidating info and posting more later today, would like to start next week, let's make February 1 the target date for rollout of the program

Quoting Rocket

I had thought previouse postings said everyon would trade all the pairs/strategies.
So whatever works, just need clarification.
  • #34
  • Quote
  • Jan 27, 2006 11:18am Jan 27, 2006 11:18am
Rocket has posted an Excel work book for our trades that looks good. I must admit that constructing Excel files like this is not a strong point for me, so I really appreciate Rocket doing this!

I would suggest everyone take a good look at it this weekend and post any comments you have so we can have it ready for use by my proposed target date of Feb 1. Also, Rocket, can you construct a weekly results work book that incorporate all five traders results? This would show results for six pairs on each of the five strategies we will be using. We could then compare strategies and pairs to get an early idea of what works best with each individual pair. After a month of this, we should have good data to work with, then build on this with fresh ideas as well as consider some BO trade testing.

Each trader can fill out their own workbook and post on the results thread that FT will put on this forum for us. That thread will be for posting results only. Our discussions will continue here.

Using the information already posted as well as ideas sent to me by PM, I am suggesting the following five strategies:

1. 10 pip inside p h/l entry
25 pip s/l
first lot t/p at median price level, second lot 50 pip t/p

2. 15 pip inside p h/l entry
45 pip s/l on gbp, 35 pip s/l on all others
90 pip t/p on gbp, 70 pip t/p on all others

3. 5 pip inside p h/l entry
30 pip s/l on gbp, 20 pip s/l on all others
t/p 20 pips inside opposite p h/l

4. 10 pip inside p h/l entry
25 pip s/l
first lot 50 pip t/p, second lot t/p at 25 inside opposite p h/l

5. 15 pip inside p h/l entry
45 pip s/l on gbp, 35 pip s/l on all others
first lot t/p at median price level, second lot t/p at 15 pips inside opposite p h/l

Let's finalize this on the weekend, so we are ready to go next week

Thanks for everyone's input and help!
  • #35
  • Quote
  • Jan 27, 2006 5:38pm Jan 27, 2006 5:38pm
Quote
QUOTE=BeachBum]That thread will be for posting results only. Our discussions will continue here.

Using the information already posted as well as ideas sent to me by PM, I am suggesting the following five strategies:

1. 10 pip inside p h/l entry
25 pip s/l
first lot t/p at median price level, second lot 50 pip t/p

2. 15 pip inside p h/l entry
45 pip s/l on gbp, 35 pip s/l on all others
90 pip t/p on gbp, 70 pip t/p on all others

3. 5 pip inside p h/l entry
30 pip s/l on gbp, 20 pip s/l on all others
t/p 20 pips inside opposite p h/l

4. 10 pip inside p h/l entry
25 pip s/l
first lot 50 pip t/p, second lot t/p at 25 inside opposite p h/l

5. 15 pip inside p h/l entry
45 pip s/l on gbp, 35 pip s/l on all others
first lot t/p at median price level, second lot t/p at 15 pips inside opposite p h/l

Let's finalize this on the weekend, so we are ready to go next week

Thanks for everyone's input and help!
This looks good as a forward testing scenario!

Due to some concerns about sample size I'd also like to suggest we begin to evaluate our results three weeks into this experiment to determine if the forward testing should continue beyond the four weeks. We can discuss this further as we go along but I have some important issues to raise about whether the testing should continue or not.

Regardless it looks as though we are almost ready to proceed. As soon as we are happy with the data collection procedure we'll be ready to log results.

In the results I'd like to see one screen shot of each intraday range for each currency, (we should standardize on either 1hr or 30 min) with the daily range plotted on the screen in a similar way to what I posted in the graphics thread of the sample system.

FT
  • #36
  • Quote
  • Jan 27, 2006 8:09pm Jan 27, 2006 8:09pm
I will make some small changes to the workbook this weekend that better groups 2 lots traded per day and also customize 5 workbooks specific to each strategy.

My metatrader demo is in 100K lot sizes but I dont see a problem.. It will be understood that I am moving the decimal place over when I log the result..
  • #37
  • Quote
  • Jan 28, 2006 12:28am Jan 28, 2006 12:28am
Quoting Rocket
I will make some small changes to the workbook this weekend that better groups 2 lots traded per day and also customize 5 workbooks specific to each strategy.

My metatrader demo is in 100K lot sizes but I dont see a problem.. It will be understood that I am moving the decimal place over when I log the result..
Why not just open a fresh account that is the same as everyone elses? It takes about 20 seconds and it does not keep you from using the other account for whatever else you opened it for?

FT
  • #38
  • Quote
  • Jan 28, 2006 8:30am Jan 28, 2006 8:30am
Quoting BeachBum
Using the information already posted as well as ideas sent to me by PM, I am suggesting the following five strategies:

1. 10 pip inside p h/l entry
25 pip s/l
first lot t/p at median price level, second lot 50 pip t/p

2. 15 pip inside p h/l entry
45 pip s/l on gbp, 35 pip s/l on all others
90 pip t/p on gbp, 70 pip t/p on all others

3. 5 pip inside p h/l entry
30 pip s/l on gbp, 20 pip s/l on all others
t/p 20 pips inside opposite p h/l

4. 10 pip inside p h/l entry
25 pip s/l
first lot 50 pip t/p, second lot t/p at 25 inside opposite p h/l

5. 15 pip inside p h/l entry
45 pip s/l on gbp, 35 pip s/l on all others
first lot t/p at median price level, second lot t/p at 15 pips inside opposite p h/l
BeachBum & others,

I think it is possible to let a fairly simple program run through historical 1 minute data that will simulate your forward testing efforts. The results from this historical dataset would provide you with a larger resultset to draw conclusions from. Slippage however wouldn't be factored in. Let me know if this is of interest and I'll look into it further.

BlueMeasure
  • #39
  • Quote
  • Jan 28, 2006 8:51am Jan 28, 2006 8:51am
I re-downloaded the file but it doesnt start a new acct. It is the same as my original.

Do I need to download from a different server than Metatrader?
Do I need to change the location it resides or rename it in any way?

How is it done?
  • #40
  • Quote
  • Jan 28, 2006 9:05am Jan 28, 2006 9:05am
Quoting Rocket
I re-downloaded the file but it doesnt start a new acct. It is the same as my original.

Do I need to download from a different server than Metatrader?
Do I need to change the location it resides or rename it in any way?

How is it done?

In MT4, under "File", choose "open a new account" at which time a new screen will pop up asking you for some personal info. Afterwards a new uid/pwd is dispatched to you - that should essentially do it.
0 traders viewing now