VantagePoint discussion thread | Page 46 | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

Daily Scalping EA - Discussion Thread 349 replies

genisis data vantagepoint 1 reply

Forward Testing Discussion Thread 241 replies

VantagePoint Sample System 0 replies

VantagePoint Graphics thread 27 replies

Attachments: VantagePoint discussion thread
Exit Attachments

VantagePoint discussion thread

  • #901
  • Jan 10, 2007 11:19am Jan 10, 2007 11:19am
Baddison,

Thank you for the better description. You always know the right words to say.


Fellow Pip Hunter
Bill
"When there is no peril in the fight there is no glory in the triumph."
  • #902
  • Jan 10, 2007 12:01pm Jan 10, 2007 12:01pm
After reading recent posts about the new version, I called VP and was told the same thing baddison has explained so eloquently. Everyone will have an opportunity to upgrade; but VP has wisely chosen to do it in stages so that their customer service department is not overwhelmed by everyone all at the same time. Makes good sense. All we need is patience until VP contacts each one of us directly because we are existing VP users. Can't wait........



Quoting baddison
gbarton13, et al....you WILL get the email. We will all get an email from VP shortly. I spoke with a rep today, and they are release the new version in stages. It make sense really...with all the new features, I would imaging the support desk will be inundated with more calls.

Any way, I was told that one of the biggest changes with the newest version is the move away from single (individual) markets to "categories". So those who purchased VP AFTER June 1st, 2006 already bought categories of markets. Those who purchase the product BEFORE June 1st, 2006 (...like myself...) bought individual markets.

For those of us who own between 69 - 89 markets, we will be offered the new version at an upgrade price of $895.
For those of us who own between 50 - 68 markets, we will be offered the new version at an upgrade price of $1495.
For those who only have one or two markets, the new version will cost around $2900 and will include many more markets than just the one or two we have.

So, the link that DavidH has posted, may not be the same link we all get...depending on how many markets we have.

Hope this helps.
  • #903
  • Jan 10, 2007 2:36pm Jan 10, 2007 2:36pm
Have placed my order to upgrade. Looking forward to the profile option (among other things) where it will reload the various markets automatically for you. A bit tiresome to go through and re-select all the markets your watching every time :-)

Been going through the webinars that were recently posted. Sheds some new light on using the product. And it looks like VP 7 adds in MACD and Stoch, two items recommended by the webinars.

David
  • #904
  • Jan 10, 2007 10:45pm Jan 10, 2007 10:45pm
Quoting Pluto9
This post is for WilliamK, BeachBum, BAddison, Stoney, and anyone else that would like to respond.

On the evening of Sunday, January 7, 2007 I used BeachBums "Set It And Forget It" System Excel spreadsheet I downloaded from this thread. Using VP's perdictions which as I interpreted per WilliamK's post 5 out of 5 of the predicted indicators agreed that I should go short on the GBP/USD entering at 1.9331 (sell) I set my sell stop at 1.9372 and set my sell limit at 1.9273. Needless to say I did not profit in this trade. I even checked the Forex Factory Calendar for any economic news that could have an impact on the market and I didn't find any.

I would really appreciate anybodys response to my post, especially those that I listed above. I am not looking to blame anybody but myself for my failures. I am only asking for some coaching or some kind of feed back. I am trying to learn from this experience so I don't repeat the same mistake again. I have had VP since March 2006, and I just can't seem to figure out a system that works for me consistently. Whatever I am doing I have been doing consistently wrong and I just don't know what it is. I don't know if it is just poor analysis of info or just bad luck. I have read all the posts on this thread. I would appreciate anybody's advise, I am here to learn. And I hope that others may learn from this as well. Thanks to all of you in this thread.

P.S.- I really believe that I have a good product in VP, and the purpose of this post is not to bring any negative light on VP but to stimulate further discussion for non-technical traders such as myself who need some more guidance and direction from the successful traders using VP.

Pluto9
Bill is right in saying that your odds are better off early in the move. On the 3rd you had a MA cross down, close alignment of the MTT and STT, a move much greater than the ATR, and the NI go negative. Those all bode well for a continuation.

With regards to order entry, since you are looking for a continuation of the move I would consider placing your entry order below the low for the previous day when shorting, above the prior day's high when going long. Make the market prove to you that it has the power to keep moving. Obviously this could possibly reduce profit but it does help protect you from reversals. Make sure you're not shorting just above support or buying just below resistance. You can get a quick glance at support/resistance levels at fx-charts.com and many other free forex sites.

Finally, remember that odds on this set-up are about 70% per Bill's posts, and he also emphasizes diversification in pairs traded. Keep trying the same set-up over and over in a demo account and see your results improve.

Best wishes,
Stoney
  • #905
  • Jan 10, 2007 11:35pm Jan 10, 2007 11:35pm
Quoting DavidH
Have placed my order to upgrade. Looking forward to the profile option (among other things) where it will reload the various markets automatically for you. A bit tiresome to go through and re-select all the markets your watching every time :-)

Been going through the webinars that were recently posted. Sheds some new light on using the product. And it looks like VP 7 adds in MACD and Stoch, two items recommended by the webinars.
Got version 7 this evening. A couple items to take note..

This version needs MS SQL Server 2005, so that gets installed (a resource hog :-( ).

Uses MSXML6, so after install, be sure to use Windows Update to get the security patches (at least for XP I needed to do this).

Does install aside of the previous VP version, so can have both 6.3 and 7.0 installed at the same time. I use CSI for my data, went to their site and downloaded a new profie that supports both VP 6.3 and 7.0, so I can still use both versions (will do some comparing of charts later one to see how different the predictions are).

This version seems to be a bit bloated when it comes to resources. The SQL server consumes almost 40 MBs just for startup. Then VP itself consumes quite a bit of resources compared to 6.3. It is real sluggish on my laptop, which is a 2.4 Ghz, 1 GIG of memory XP machine. On my main machine, a 3.2 Ghz hyperthreaded, 4 GIGs of memory, 4 monitor, XP machine it works okay (a little doggish when I have VP 7 and Tradestation open though). It is kind of cool to be able to stretch VP across all 4 screens, can get quite a few charts up. The profile feature is very nice!

Will be playing with it the next few days.

David
  • #906
  • Jan 11, 2007 12:18am Jan 11, 2007 12:18am
I also purchased it this evening and had to go through all the hassles of installing Windows XP SP1 & 2 packs along with SQL Server. Once I finally got it all working, I must say I am impressed with the new version. I will elaborate on that a bit further tomorrow after VP's prediction for the day is over and it would be ok to share todays prediction.
  • #907
  • Edited 3:45pm Jan 11, 2007 2:00pm | Edited 3:45pm
Hi all,
I just started reading this thread last night, and thought I would join. My partner and I have been trading Forex with Vantage Point for the past year with a tremendous success, and we have been using a beta version of 7.0 for about a month.

Here are my notes on the new indicators and the improved performance.

There is a new long term indicator to go with the short and medium term indicators. Also the strength indicator was changed to be a predicted very short (3 day) term indicator that follows the neural index exactly. (so if you use the 6.4 strength indicator in your trading, then be advised it will now respond differently) Both of the new predicted average indicators work well with perfect order strategies. There are three new momentum indicators, each predicted one day ahead. They work great, but for some reason, they are not yet available for export so back testing with the momentum indicators has to be done manually. They did change the history export to provide 5 years of back data, but beware you can slow everything to standstill if you try to open too many history files at once.

The big news is that they increased the number of markets that are used for intermarket prediction from 9 to 25 and now calculate using the midpoint between the high and the low for the day rather than the close of the day(which is meaningless in the forex market.) I did an accuracy test on the all the FOREX markets for the past 2 years and got an overall accuracy of 81.6% for all forex markets combined. (6.4 gives 74.3% by the same metric)

As for trading strategies, we found that the best way to use Vantage Point is to match the VP indicators (or any combination of indicators) with either a trade entrance, exit, or stop loss strategy for a given currency pair. Markets change, VP indicators do not, so find a setup that has worked very well for at least six months (12 months is better) with a certain currency pair and use it as long as it performs to a level acceptable to you. (We look for trade strategies that give a 10/1 win loss ratio (WLR = Pips won/Pips Lost) and only trade strategies with better than a 5/1 WLR.

I am posting one of our strategies here that we published elsewhere. I developed it to see if I could find a strategy that would take advantage of the new stochastice indicator. In this case, I use the stochastics predicted one day ahead as an advance notice of over bought or over sold, I use it as an exit strategy in a range trading market.

If you decide to try this strategy, please realize that I developed it for the EURCHF market only, and every rule in the strategy was added to maximize the WLR. I will only trade this on the EURCHF, and I will discard or modify it when it falls below a 12 month 5/1 WLR.

EURCHF

  1. Enter at the “close” price when the “Predicted Medium Term Difference” crosses zero in the direction of the trade, and when the “Predicted Stochastic” is between 30-60 for long and 40-70 for short.
  2. Set and Reset the stop loss daily to right at the “Predicted Next Day High” or “Low” (whichever is applicable).
  3. Exit at the “close” price when the “Predicted Stochastic” moves above 70 for long or below 30 for short or when either the “Predicted Next Day High” or the “Predicted Next Day Low” predicts the next day in a direction against the trade

Here are the 12 month statistics on it. Please note this trade setup only occured 8 times in 12 months.
<TABLE class=MsoNormalTable style="MARGIN: auto auto auto 4.65pt; WIDTH: 373.35pt; BORDER-COLLAPSE: collapse; mso-padding-alt: 0in 5.4pt 0in 5.4pt" cellSpacing=0 cellPadding=0 width=498 border=0><TBODY><TR style="HEIGHT: 12.75pt; mso-yfti-irow: 0"><TD style="BORDER-RIGHT: #ece9d8; PADDING-RIGHT: 5.4pt; BORDER-TOP: #ece9d8; PADDING-LEFT: 5.4pt; PADDING-BOTTOM: 0in; BORDER-LEFT: #ece9d8; WIDTH: 76.2pt; PADDING-TOP: 0in; BORDER-BOTTOM: #ece9d8; HEIGHT: 12.75pt; BACKGROUND-COLOR: transparent" vAlign=bottom noWrap width=102>Starting Date<?xml:namespace prefix = o ns = "urn:schemas-microsoft-com:office:office" /><o></o>





</TD><TD style="BORDER-RIGHT: #ece9d8; PADDING-RIGHT: 5.4pt; BORDER-TOP: #ece9d8; PADDING-LEFT: 5.4pt; PADDING-BOTTOM: 0in; BORDER-LEFT: #ece9d8; WIDTH: 71.95pt; PADDING-TOP: 0in; BORDER-BOTTOM: #ece9d8; HEIGHT: 12.75pt; BACKGROUND-COLOR: transparent" vAlign=bottom noWrap width=96>Ending Date<o></o>





</TD><TD style="BORDER-RIGHT: #ece9d8; PADDING-RIGHT: 5.4pt; BORDER-TOP: #ece9d8; PADDING-LEFT: 5.4pt; PADDING-BOTTOM: 0in; BORDER-LEFT: #ece9d8; WIDTH: 46.1pt; PADDING-TOP: 0in; BORDER-BOTTOM: #ece9d8; HEIGHT: 12.75pt; BACKGROUND-COLOR: transparent" vAlign=bottom noWrap width=61># wins<o></o>





</TD><TD style="BORDER-RIGHT: #ece9d8; PADDING-RIGHT: 5.4pt; BORDER-TOP: #ece9d8; PADDING-LEFT: 5.4pt; PADDING-BOTTOM: 0in; BORDER-LEFT: #ece9d8; WIDTH: 24.95pt; PADDING-TOP: 0in; BORDER-BOTTOM: #ece9d8; HEIGHT: 12.75pt; BACKGROUND-COLOR: transparent" vAlign=bottom noWrap width=33>Pips <o></o>





</TD><TD style="BORDER-RIGHT: #ece9d8; PADDING-RIGHT: 5.4pt; BORDER-TOP: #ece9d8; PADDING-LEFT: 5.4pt; PADDING-BOTTOM: 0in; BORDER-LEFT: #ece9d8; WIDTH: 48.65pt; PADDING-TOP: 0in; BORDER-BOTTOM: #ece9d8; HEIGHT: 12.75pt; BACKGROUND-COLOR: transparent" vAlign=bottom noWrap width=65># losses<o></o>





</TD><TD style="BORDER-RIGHT: #ece9d8; PADDING-RIGHT: 5.4pt; BORDER-TOP: #ece9d8; PADDING-LEFT: 5.4pt; PADDING-BOTTOM: 0in; BORDER-LEFT: #ece9d8; WIDTH: 48pt; PADDING-TOP: 0in; BORDER-BOTTOM: #ece9d8; HEIGHT: 12.75pt; BACKGROUND-COLOR: transparent" vAlign=bottom noWrap width=64>Pips<o></o>





</TD><TD style="BORDER-RIGHT: #ece9d8; PADDING-RIGHT: 5.4pt; BORDER-TOP: #ece9d8; PADDING-LEFT: 5.4pt; PADDING-BOTTOM: 0in; BORDER-LEFT: #ece9d8; WIDTH: 57.5pt; PADDING-TOP: 0in; BORDER-BOTTOM: #ece9d8; HEIGHT: 12.75pt; BACKGROUND-COLOR: transparent" vAlign=bottom noWrap width=77>Win/Loss<o></o>





</TD></TR><TR style="HEIGHT: 12.75pt; mso-yfti-irow: 1"><TD style="BORDER-RIGHT: #ece9d8; PADDING-RIGHT: 5.4pt; BORDER-TOP: #ece9d8; PADDING-LEFT: 5.4pt; PADDING-BOTTOM: 0in; BORDER-LEFT: #ece9d8; WIDTH: 76.2pt; PADDING-TOP: 0in; BORDER-BOTTOM: #ece9d8; HEIGHT: 12.75pt; BACKGROUND-COLOR: transparent" vAlign=bottom noWrap width=102><o></o>





</TD><TD style="BORDER-RIGHT: #ece9d8; PADDING-RIGHT: 5.4pt; BORDER-TOP: #ece9d8; PADDING-LEFT: 5.4pt; PADDING-BOTTOM: 0in; BORDER-LEFT: #ece9d8; WIDTH: 71.95pt; PADDING-TOP: 0in; BORDER-BOTTOM: #ece9d8; HEIGHT: 12.75pt; BACKGROUND-COLOR: transparent" vAlign=bottom noWrap width=96><o></o>





</TD><TD style="BORDER-RIGHT: #ece9d8; PADDING-RIGHT: 5.4pt; BORDER-TOP: #ece9d8; PADDING-LEFT: 5.4pt; PADDING-BOTTOM: 0in; BORDER-LEFT: #ece9d8; WIDTH: 46.1pt; PADDING-TOP: 0in; BORDER-BOTTOM: #ece9d8; HEIGHT: 12.75pt; BACKGROUND-COLOR: transparent" vAlign=bottom noWrap width=61><o></o>





</TD><TD style="BORDER-RIGHT: #ece9d8; PADDING-RIGHT: 5.4pt; BORDER-TOP: #ece9d8; PADDING-LEFT: 5.4pt; PADDING-BOTTOM: 0in; BORDER-LEFT: #ece9d8; WIDTH: 24.95pt; PADDING-TOP: 0in; BORDER-BOTTOM: #ece9d8; HEIGHT: 12.75pt; BACKGROUND-COLOR: transparent" vAlign=bottom noWrap width=33>won<o></o>





</TD><TD style="BORDER-RIGHT: #ece9d8; PADDING-RIGHT: 5.4pt; BORDER-TOP: #ece9d8; PADDING-LEFT: 5.4pt; PADDING-BOTTOM: 0in; BORDER-LEFT: #ece9d8; WIDTH: 48.65pt; PADDING-TOP: 0in; BORDER-BOTTOM: #ece9d8; HEIGHT: 12.75pt; BACKGROUND-COLOR: transparent" vAlign=bottom noWrap width=65><o></o>





</TD><TD style="BORDER-RIGHT: #ece9d8; PADDING-RIGHT: 5.4pt; BORDER-TOP: #ece9d8; PADDING-LEFT: 5.4pt; PADDING-BOTTOM: 0in; BORDER-LEFT: #ece9d8; WIDTH: 48pt; PADDING-TOP: 0in; BORDER-BOTTOM: #ece9d8; HEIGHT: 12.75pt; BACKGROUND-COLOR: transparent" vAlign=bottom noWrap width=64>Lost<o></o>





</TD><TD style="BORDER-RIGHT: #ece9d8; PADDING-RIGHT: 5.4pt; BORDER-TOP: #ece9d8; PADDING-LEFT: 5.4pt; PADDING-BOTTOM: 0in; BORDER-LEFT: #ece9d8; WIDTH: 57.5pt; PADDING-TOP: 0in; BORDER-BOTTOM: #ece9d8; HEIGHT: 12.75pt; BACKGROUND-COLOR: transparent" vAlign=bottom noWrap width=77>Ratio<o></o>





</TD></TR><TR style="HEIGHT: 12.75pt; mso-yfti-irow: 2; mso-yfti-lastrow: yes"><TD style="BORDER-RIGHT: #ece9d8; PADDING-RIGHT: 5.4pt; BORDER-TOP: #ece9d8; PADDING-LEFT: 5.4pt; PADDING-BOTTOM: 0in; BORDER-LEFT: #ece9d8; WIDTH: 76.2pt; PADDING-TOP: 0in; BORDER-BOTTOM: #ece9d8; HEIGHT: 12.75pt; BACKGROUND-COLOR: transparent" vAlign=bottom noWrap width=102>

<?xml:namespace prefix = st1 ns = "urn:schemas-microsoft-com:office:smarttags" /><st1:date Month="1" Day="1" Year="2006">1/1/2006</st1:date><o></o>










</TD><TD style="BORDER-RIGHT: #ece9d8; PADDING-RIGHT: 5.4pt; BORDER-TOP: #ece9d8; PADDING-LEFT: 5.4pt; PADDING-BOTTOM: 0in; BORDER-LEFT: #ece9d8; WIDTH: 71.95pt; PADDING-TOP: 0in; BORDER-BOTTOM: #ece9d8; HEIGHT: 12.75pt; BACKGROUND-COLOR: transparent" vAlign=bottom noWrap width=96>

<st1:date Month="12" Day="18" Year="2006">12/18/2006</st1:date><o></o>










</TD><TD style="BORDER-RIGHT: #ece9d8; PADDING-RIGHT: 5.4pt; BORDER-TOP: #ece9d8; PADDING-LEFT: 5.4pt; PADDING-BOTTOM: 0in; BORDER-LEFT: #ece9d8; WIDTH: 46.1pt; PADDING-TOP: 0in; BORDER-BOTTOM: #ece9d8; HEIGHT: 12.75pt; BACKGROUND-COLOR: transparent" vAlign=bottom noWrap width=61>

6<o></o>










</TD><TD style="BORDER-RIGHT: #ece9d8; PADDING-RIGHT: 5.4pt; BORDER-TOP: #ece9d8; PADDING-LEFT: 5.4pt; PADDING-BOTTOM: 0in; BORDER-LEFT: #ece9d8; WIDTH: 24.95pt; PADDING-TOP: 0in; BORDER-BOTTOM: #ece9d8; HEIGHT: 12.75pt; BACKGROUND-COLOR: transparent" vAlign=bottom noWrap width=33>

358<o></o>










</TD><TD style="BORDER-RIGHT: #ece9d8; PADDING-RIGHT: 5.4pt; BORDER-TOP: #ece9d8; PADDING-LEFT: 5.4pt; PADDING-BOTTOM: 0in; BORDER-LEFT: #ece9d8; WIDTH: 48.65pt; PADDING-TOP: 0in; BORDER-BOTTOM: #ece9d8; HEIGHT: 12.75pt; BACKGROUND-COLOR: transparent" vAlign=bottom noWrap width=65>

2<o></o>










</TD><TD style="BORDER-RIGHT: #ece9d8; PADDING-RIGHT: 5.4pt; BORDER-TOP: #ece9d8; PADDING-LEFT: 5.4pt; PADDING-BOTTOM: 0in; BORDER-LEFT: #ece9d8; WIDTH: 48pt; PADDING-TOP: 0in; BORDER-BOTTOM: #ece9d8; HEIGHT: 12.75pt; BACKGROUND-COLOR: transparent" vAlign=bottom noWrap width=64>

-19<o></o>










</TD><TD style="BORDER-RIGHT: #ece9d8; PADDING-RIGHT: 5.4pt; BORDER-TOP: #ece9d8; PADDING-LEFT: 5.4pt; PADDING-BOTTOM: 0in; BORDER-LEFT: #ece9d8; WIDTH: 57.5pt; PADDING-TOP: 0in; BORDER-BOTTOM: #ece9d8; HEIGHT: 12.75pt; BACKGROUND-COLOR: transparent" vAlign=bottom noWrap width=77>

18.8<o></o>










</TD></TR></TBODY></TABLE>
  • #908
  • Jan 11, 2007 3:07pm Jan 11, 2007 3:07pm
Thanks for sharing with us MrSmiley, this deserves a careful look!

Quoting MrSmiley
Hi all,
I just started reading thread this last night, and thought I would join. My partner and I have been trading Forex with Vantage Point for the past year with a tremendous success, and we have been using a beta version of 7.0 for about a month.
  • #909
  • Jan 11, 2007 9:36pm Jan 11, 2007 9:36pm
Mr Smiley,

Very nice work. I have 1 question 4 now. When U talk about the Predicted Stochs what # do U look @ ,this is 7.0, PStoch or Trigger ?

Fellow Pip Hunter

Bill
"When there is no peril in the fight there is no glory in the triumph."
  • #910
  • Jan 12, 2007 1:05am Jan 12, 2007 1:05am
Thanks Bill,

The trade calls for the PStoch not the trigger. See the attachment. (This was the one time when two trades occured close enough in time to see them on the same chart.)

Steve
Attached Image
  • #911
  • Jan 12, 2007 12:14pm Jan 12, 2007 12:14pm
Hi All

I just went to http://www.traderslibrary.com looking for information on backtesting.

Perhaps some of you that have looked into this yourselfs; could give the rest of us a good starting place to educate our selfs.

I hope some of you would be will be willing to help the rest of us refine our trading decisions, by posting sources for Backtesting.


Thanks

Cyclone
Cyclone
  • #912
  • Jan 12, 2007 11:14pm Jan 12, 2007 11:14pm
Interesting you should ask this. I read a very good book called getting started in technical analysis by Jack D. Schwager and he has you back testing any strategy back two years over more than one market to account for market changes.

I spent six months and counless hours testing vantage point solutions with this standard of testing, and I did not find any of the types of high performance trades I knew vantage point was capable of producing. What I finally realized was that since vantage point is actively predicting the market using neural networks and inter market analysis, the indicators should be treated differently than normal passive indicators that react to price alone.

What I have found is that each vantage point indicator best predicts markets that have a certain voulatility. Markets are always changing. Think of each vantage point indicator as having a "sweet spot". When a given changing market slides past the "sweet spot" of a vantage point indicator, then that vantage point indicator starts predicting the market with what sometimes appears to be uncanny accuracy. Once I realized this, I started testing for it, and I found that there were lots of trading scinarios that were amazingly accurate for 9 to 18 months(sometimes more) and then market conditions would gradually change and they would lose their effectiveness. Often in the middle of the cycle there can be a huge win/loss ratio(WLR). So we started looking for trade scinarios with a 6 to 12 month trading record of better than 10/1 (WLR)(total pips won/total pips lost) and then we trade these scinarios assuming that they will still be effective for a while, but expect them to slowly lose their effectivness. We are looking to keep our actual trading above 5/1. The trade I showed earlier is at >18/1 and is one of our current trades. You are only limited by your imagination and your ingenuity to match vantage point indicators (single or combined) to a market. What you are looking for are indicators that best predict the three components you need for a trade, namely the entry, exit and daily stop loss (or profit protection). You can use different indicators or combinations of indicators for each trade component.

For my back testing, I enter the trade at the historical close of day price and play the scinario out. for those who have never done back testing you simply play act the trade as you would have had you recieved the daily vantage point data real time. It is best to program the backtesting using excel or something similar to get an unbiased answer, but you can do it by hand. Beware, if you do it by hand, you have to be brutally honest in writing down the trades or you will end up with falsly good data and find that out when you start trading the scinario.
  • #913
  • Jan 13, 2007 5:15pm Jan 13, 2007 5:15pm
I have used a system I developed with VP.
It is a buy/forget system.

Basically, a 2 period Moving average is constructed.
For Longs.
If the close is less than the 2MA and NN is 1 and the predicted MA5>MA10, then go long.

A target profit is set and a stop loss is set.
Additionally, a band is set and an offset selected for entry.

The TP, SL , band , and offset have been selected as a function of the
day of week.

I am going to attached the Tradestation system testing files for
both the GBPUSD and EURUSD pairs. (the detailed settings are included
in the tradestation report)

For optimal profits the win percents are in the 60% range, and both
systems generate about 8K per year.

What I would like to see is a history file from the new version 7.0 for either of these pairs.

enjoy
wart
Attached File(s)
File Type: zip EURUSDnn.zip   576 KB | 303 downloads
File Type: zip cablenn.zip   512 KB | 253 downloads
  • #914
  • Edited 7:31pm Jan 14, 2007 7:19pm | Edited 7:31pm
Quoting MrSmiley
I did an accuracy test on the all the FOREX markets for the past 2 years and got an overall accuracy of 81.6% for all forex markets combined. (6.4 gives 74.3% by the same metric)
Could you define your backtesting method?
I have done a lot of backtesting using MT4 and exported VP data - minimum 2 years backtest data. However, I have to say the accuracy does not reach those levels. These were tests based on how you should use VP though.
I did come up with one strategy that works only on one pair that is about 90% accurate with virtually 0 drawdown however only about 10 trades per year...
thanks
  • #915
  • Jan 15, 2007 12:21am Jan 15, 2007 12:21am
Hi Warthog

Here are a couple of 7.0 history files. Your short trade performance on the GBPUSD looks fantastic especially for a buy/forget! What I have been calling the win/loss ratio looks equilivent to the trade station profit factor. (gross profit/gross loss)

It is more work, but I find that most trades can benifit from using Vantage Point intelligence to call a trade rather setting a strict limit target. I did quite a bit if testing on the GBPUSD with 6.4. Try testing what your track record would have been if you had set you stop loss each day to the next day's predicted high of low, and exited either when you stopped out or at the market close price when the 6.4 strength indicator crossed zero in a direction against the trade. BTW. The EURUSD probably does not respond as well to thise two indicators.

Steve
Attached File(s)
File Type: zip History files.zip   286 KB | 289 downloads
  • #916
  • Jan 15, 2007 1:31am Jan 15, 2007 1:31am
Quoting zarni
Could you define your backtesting method?
Hi Zarni,
The Neural index is defined by the 3 day simple moving average two days out. In 7.0 it is based in the midpoint.

Take the midpoint price between the high and the low for each of the past three days including today add them up and divide by three. That is the three day simple moving average.

The Neural Indicator tells you if (Today's midpoint + Tomorrow + the next day)/3 will be greater or less than
(Today's midpoint + Yesterday + the day before)/3

That is what I measured when I got an overall accuracy of 81.6% for all forex markets combined over a two year historical period. (6.4 gives 74.3% by the same metric)

This does not mean you can trade the neural index alone. It takes more to craft a trade than the neural index. However if you found a really great trade that trades 10 times a year, great! Find about ten more of those, and you will have an amazing trading system!

One word of caution, test your trade back as far as you can go to see how it fails as the market changes. In 7.0 you get 5 years of historical back data. And just as you found that that trade only worked in one market, I found that you should expect that market to change and for that trade to eventually lose it's effectiveness. Based on our back testing of this, we look for trades that have a good year of track record, and we expect them to still be good for at least another 6 months minimum.

We look for trades that have a better than 10 to 1 Pips (or dollars) won for every pip lost and trade them untill their 12 month performance falls below 5 to 1. Most of our best trades trade about once a month like the one you found. To get it that high and higher, you have to spend as much time crafting your exit strategy with Vantage Point indicators as you do crafting your entrance strategy.

Hope this helps, this is essentially the learning path that we have been on this past year. The real secret I believe to trading with neural networks, is that each indicator works best on markets with a certain voulitiality. Since markets are always changing, find out when a market and an indicator are lining up, and use it while it works.
  • #917
  • Jan 15, 2007 1:39am Jan 15, 2007 1:39am
Hi,

Thanks for the v7.0 history files.

At first glance, the historical nueral index is different between V7
and my version 6.4.

Additionally, the medium term and short term trend values are also
different.

now the questions,
1. Can I continue to trust V6.4 predictions?
2. Am I forced to upgrade?



thanks again
  • #918
  • Jan 15, 2007 7:10am Jan 15, 2007 7:10am
Quoting MrSmiley
Hi Zarni,
The Neural index is defined by the 3 day simple moving average two days out. In 7.0 it is based in the midpoint.

Take the midpoint price between the high and the low for each of the past three days including today add them up and divide by three. That is the three day simple moving average.

The Neural Indicator tells you if (Today's midpoint + Tomorrow + the next day)/3 will be greater or less than
(Today's midpoint + Yesterday + the day before)/3

That is what I measured when I got an overall accuracy of 81.6% for all forex markets combined over a two year historical period. (6.4 gives 74.3% by the same metric)

This does not mean you can trade the neural index alone. It takes more to craft a trade than the neural index. However if you found a really great trade that trades 10 times a year, great! Find about ten more of those, and you will have an amazing trading system!

One word of caution, test your trade back as far as you can go to see how it fails as the market changes. In 7.0 you get 5 years of historical back data. And just as you found that that trade only worked in one market, I found that you should expect that market to change and for that trade to eventually lose it's effectiveness. Based on our back testing of this, we look for trades that have a good year of track record, and we expect them to still be good for at least another 6 months minimum.

We look for trades that have a better than 10 to 1 Pips (or dollars) won for every pip lost and trade them untill their 12 month performance falls below 5 to 1. Most of our best trades trade about once a month like the one you found. To get it that high and higher, you have to spend as much time crafting your exit strategy with Vantage Point indicators as you do crafting your entrance strategy.

Hope this helps, this is essentially the learning path that we have been on this past year. The real secret I believe to trading with neural networks, is that each indicator works best on markets with a certain voulitiality. Since markets are always changing, find out when a market and an indicator are lining up, and use it while it works.
Hi there,
thanks for reply.
yes the neural index I also found to be a very accurate indicator (same as what is claimed). However, I have to say at least to me it is virtually useless.
Here is the reason - the neural index was on 6.4 at least based on average of yesterday and day before compared to tomorrow and day after. So it stands to reason that if today was already higher than yesterday and day before that tomorrow and day after would be very high probability would be higher. I tried numerous ways to take advantage of this but could not for the life of me put this to any good use. Backtesting of numerous ways to use that indicator showed breakeven or slightly better results on balance. I had more success with the other indicators.
-martin
  • #919
  • Jan 15, 2007 8:44am Jan 15, 2007 8:44am
hi i have just bought vp i am new to trading and need a system or simple set of rules to use with the vp software,
apparently i can not just use it on its own to make trades pay,
any ideas?
thanks
b
  • #920
  • Jan 15, 2007 1:19pm Jan 15, 2007 1:19pm
Quoting warthog
now the questions,
1. Can I continue to trust V6.4 predictions?
2. Am I forced to upgrade?
Your 6.4 version runs independently, it doesn't change because there is a new version so it will continue it's same performance.

The 6.4 neural index is calculated from an intermarket analysis of 9 markets The 7.0 neural index is calculated from an intermarket analysis of 25 markets. Based in my calculations, the 74.3% accuracy for 6.4 means that 26 out of 100 predictions are wrong. The 7.0 accuracy of 81.6% means that 18 out of 100 predictions are wrong. One way to think of this, is that 7.0 has about 30% less neural index failures than 6.4.

Steve
6 traders viewing now