VantagePoint discussion thread | Page 31 | 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

  • #601
  • Jul 18, 2006 4:25am Jul 18, 2006 4:25am
CP, Zarni,

I think I've found a flaw on the exit. After the cross signal to exit, you exit at the closing price for the same day, but by the time we have the data the closing price has gone.

Please let me know if I've missed something here.
  • #602
  • Jul 18, 2006 7:01am Jul 18, 2006 7:01am
Quoting rodant
CP, Zarni,

I think I've found a flaw on the exit. After the cross signal to exit, you exit at the closing price for the same day, but by the time we have the data the closing price has gone.

Please let me know if I've missed something here.
Yes backtesting on a daily bar with signals that come in at an hour or so past the open it is really innacurate.
The only thing you can really do is set the trade times to around 6:30PM EST when you get the vp data and then backtest on hourly. but as was already mentioned you can get a good feel for which markets work better by just eyeballing them. gold does tend to trend more...
  • #603
  • Jul 18, 2006 10:51am Jul 18, 2006 10:51am
Quoting rodant
CP, Zarni,

I think I've found a flaw on the exit. After the cross signal to exit, you exit at the closing price for the same day, but by the time we have the data the closing price has gone.

Please let me know if I've missed something here.
Yes it's true, that the data comes about an hour after the closing price has gone. That is one of the limitations of testing with this end of day data. However, it's also one of the least active times of the market for most of the currencies most of the time. So I think it would for the most part average out to be about the same and makes a rather minor difference overall. Also, never view backtesting as an accurate result of what you will actually achieve with any system. Backtesting is more of giving you an idea of whether or not a system has potential to be profitable.

CP
Making my living one pip at a time.....
  • #604
  • Jul 28, 2006 2:14am Jul 28, 2006 2:14am
Quoting BeachBum
Hello MW...

Here are a few observations based on much research and use of VP:

IMO you will find very few occasions when all the indicators you list are coherent, meaning few trades. When the market is ranging, I don't think you will reach your profit target. And when the market is trending, you won't hit the entry point to trigger the trade. This is because the accuracy of VP daily predictions varies according to market conditions: greater accuracy when ranging, less accuracy when trending. And finally, if you do enter a trade in a ranging market, your stop loss should be outside of the predicted range, otherwise it is very likely you will get stopped out.

However, I think you should forward test your strategy and share the results. Or you might want to do some backtesting with one or two currency pairs. We all learn from the experiences and work done by others here at FF.
Hi BB

I am a new user of VP (few days old). I am kind a interested with the phrase: "This is because the accuracy of VP daily predictions varies according to market conditions: greater accuracy when ranging, less accuracy when trending". I would appreciate if you could elaborate further what is ranging market and what is trending market. Pretty please!!!
  • #605
  • Jul 28, 2006 8:41am Jul 28, 2006 8:41am
Quoting maxjbd1024
Hi BB

I am a new user of VP (few days old). I am kind a interested with the phrase: "This is because the accuracy of VP daily predictions varies according to market conditions: greater accuracy when ranging, less accuracy when trending". I would appreciate if you could elaborate further what is ranging market and what is trending market. Pretty please!!!
here's an article that should help explain:

http://www.investopedia.com/articles.../05/050505.asp
  • #606
  • Jul 28, 2006 9:26am Jul 28, 2006 9:26am
Is VP comming with 30 days money back?
Like everything?
Try hard, think fast, die young
  • #607
  • Jul 29, 2006 7:49am Jul 29, 2006 7:49am
No money back guarantee
  • #608
  • Jul 30, 2006 7:49pm Jul 30, 2006 7:49pm
Quoting BeachBum
here's an article that should help explain:

http://www.investopedia.com/articles.../05/050505.asp
thnx BB
  • #609
  • Aug 1, 2006 1:42pm Aug 1, 2006 1:42pm
I just bought VantagePoint. Apparently I didn't understand something correctly. When I bought one market I thought I was buing the Forex market (all currency pairs) but apprently I'm only going to get one currency pair. I'll try to talk them into more but I don't know if they'll just give them to me since I already paid for the software. Oh well, my question is which currency pair do you all recommend I get?

I was thinking of GBP/CHF, GBP/JPY, or EUR/USD. If I only get one choice, which one do you think would be the most profiable for me to choose with VP?


Thanks,
~Keith
  • #610
  • Aug 1, 2006 1:49pm Aug 1, 2006 1:49pm
Quoting floridaforex
I just bought VantagePoint. Apparently I didn't understand something correctly. When I bought one market I thought I was buing the Forex market (all currency pairs) but apprently I'm only going to get one currency pair. I'll try to talk them into more but I don't know if they'll just give them to me since I already paid for the software. Oh well, my question is which currency pair do you all recommend I get?

I was thinking of GBP/CHF, GBP/JPY, or EUR/USD. If I only get one choice, which one do you think would be the most profiable for me to choose with VP?


Thanks,
~Keith
There is no right or wrong answer here. It all comes down to your preference. Generally pick a higher volatility, higher liquidity pair with relatively low spreads with your broker.
Cable is usually my choice but that is just my preference.
  • #611
  • Aug 2, 2006 11:31am Aug 2, 2006 11:31am
Thanks Zarni. I just talked to my rep at VP and it ends up that I subscribed to all 14 forex pairs like I originally thought. The way the software warned about the experation made it sound like each pair needed to be purchased seperatly.

What a relief!!
  • #612
  • Aug 5, 2006 12:55am Aug 5, 2006 12:55am
Does anyone know where I can get ahold of some good intraday historical data?

I have built spreadsheet in Excel that can import data from the VP databases. My calculations showed that for the EUR/USD and GBP/CHF (I haven't tried any others yet) that VP accurately predicts the median about 89% of the time.
Sometimes VP predicts the correct direction, but the price doesn't actually reach the median. In this case there may still be a profit even though it didn't hit median. This increases the profitable trades to about 93%.

**When I say it predicts the median correctly what I mean is that sometime during the day the market price will hit the predicted median price.

Assuming that by the time I get the data from CSI (6:30) the market hasn't already hit the median, I enter the market at the current price with a limit of the median price. If this is a short, I would set the stop limit to the predicted high, if it's a long position then the stop would be the predicted low.

Backtesting shows that the maximum loss in GBP/CHF was about 350 pips in one day. If you never took the money out of the account, the account would have had a max drawdown of 160 pips. After 51 days of trading it never had drawdown again.
The EUR/USD had a max loss of about 233 pips in one day, no drawdown.

Total profits calculated from as far back as my VP database provides:

GBP/CHF: 24011 93% profitable trades (since 1993)
EUR/USD: 5361 91% profitable trades (since 1996)

The accuracy rates remain around 90% no matter what time period I look at. The only problem with this is that by the time I get the data the close price is gone so I can't enter there all the time. If someone knows of a place where I could get intraday historical data I could reference the price at say 6:40 EST and base the entry on that price. It would let me know how many of the trades are even possible, and if it would still be a profitable strategy.

Also, if the price overshot PH/PL (in opposite direction of trade) before it hit median the trade would have been stopped out. Without intraday data I have no way of seeing which trades would have been stopped out. If I can't get a hold of intraday data I may just be conservative and calculate overshots as thought they happend before the median was hit and count those trades as losses so the results are more realistic.

Once I am able to get this to a useable state, or successfully forward test it I will upload the file. If anyone wants to see it now just let me know and I'll go ahead and post it on here. I've only been working on it for a couple days, so there's a possibility I miscalculated or left something out but I 'm pretty sure I've worked out the kinks.

Thanks in advance,
~Keith
  • #613
  • Aug 5, 2006 8:02am Aug 5, 2006 8:02am
Hello FF........

I'm interested in what you are doing. Isn't the information you want available from Metatrader?

I use CSI data feed which is available most days at 5:15 ET. Would this earlier time help your strategy?

And finally: most days at 6 PM ET, the current price on most pairs is very close to the median (many times it has been hit between 5 PM and 6 PM). How do you plan to trade this scenario? I am developing a strategy that allows me to trade "away" from the median "area" toward the p/h or p/l.

Looking forward to your comments..........

BB


Quoting floridaforex
Does anyone know where I can get ahold of some good intraday historical data?

I have built spreadsheet in Excel that can import data from the VP databases. My calculations showed that for the EUR/USD and GBP/CHF (I haven't tried any others yet) that VP accurately predicts the median about 89% of the time.
Sometimes VP predicts the correct direction, but the price doesn't actually reach the median. In this case there may still be a profit even though it didn't hit median. This increases the profitable trades to about 93%.

**When I say it predicts the median correctly what I mean is that sometime during the day the market price will hit the predicted median price.

Assuming that by the time I get the data from CSI (6:30) the market hasn't already hit the median, I enter the market at the current price with a limit of the median price. If this is a short, I would set the stop limit to the predicted high, if it's a long position then the stop would be the predicted low.

Backtesting shows that the maximum loss in GBP/CHF was about 350 pips in one day. If you never took the money out of the account, the account would have had a max drawdown of 160 pips. After 51 days of trading it never had drawdown again.
The EUR/USD had a max loss of about 233 pips in one day, no drawdown.

Total profits calculated from as far back as my VP database provides:

GBP/CHF: 24011 93% profitable trades (since 1993)
EUR/USD: 5361 91% profitable trades (since 1996)

The accuracy rates remain around 90% no matter what time period I look at. The only problem with this is that by the time I get the data the close price is gone so I can't enter there all the time. If someone knows of a place where I could get intraday historical data I could reference the price at say 6:40 EST and base the entry on that price. It would let me know how many of the trades are even possible, and if it would still be a profitable strategy.

Also, if the price overshot PH/PL (in opposite direction of trade) before it hit median the trade would have been stopped out. Without intraday data I have no way of seeing which trades would have been stopped out. If I can't get a hold of intraday data I may just be conservative and calculate overshots as thought they happend before the median was hit and count those trades as losses so the results are more realistic.

Once I am able to get this to a useable state, or successfully forward test it I will upload the file. If anyone wants to see it now just let me know and I'll go ahead and post it on here. I've only been working on it for a couple days, so there's a possibility I miscalculated or left something out but I 'm pretty sure I've worked out the kinks.

Thanks in advance,
~Keith
  • #614
  • Aug 5, 2006 8:29am Aug 5, 2006 8:29am
Quoting BeachBum
Hello FF........

I'm interested in what you are doing. Isn't the information you want available from Metatrader?

I use CSI data feed which is available most days at 5:15 ET. Would this earlier time help your strategy?

And finally: most days at 6 PM ET, the current price on most pairs is very close to the median (many times it has been hit between 5 PM and 6 PM). How do you plan to trade this scenario? I am developing a strategy that allows me to trade "away" from the median "area" toward the p/h or p/l.

Looking forward to your comments..........

BB
Hi BB,
i am wondering how you get 5:15 for csi. I am never able to get it anywhere near that. The best time I've managed is 5:30 and that is really rate. Usually it is about 6 or 6:30 EST.
  • #615
  • Aug 5, 2006 8:37am Aug 5, 2006 8:37am
Zarni...........

I don't know why the difference. I just logon and download everyday at 5:15 ET. Occasionally the data is a little later, perhaps 5:30, but something is wrong if you're not getting it until 6 or 6:30. Genesis does not make their data available until 6:30 ET, which is much later than I want, in part because I am interested some days in trading "away" from the p/m price. You might want to contact CSI about this. They once told me that the Forex numbers were actually in their system by 4:45 ET but not available for download until after 5 PM.


Quoting zarni
Hi BB,
i am wondering how you get 5:15 for csi. I am never able to get it anywhere near that. The best time I've managed is 5:30 and that is really rate. Usually it is about 6 or 6:30 EST.
  • #616
  • Aug 5, 2006 3:25pm Aug 5, 2006 3:25pm
That is one of the main problems with my system. Because I get the data so late I cannot usually trade it, so it's only good in theory.

I haven't been able to get my data before 6:30. Yesterday I tried downloading it at 6:25 and it wasn't updated yet. I tried again 5 minutes later and it worked. I'd love to get my data at 5:15.

I just got a hold of hourly intraday data going back to Jan05. I'll let you know what I find out.

Do you get your data at 5 EST or central time? I'm never able to get it that early and I use CSI.
  • #617
  • Aug 5, 2006 6:22pm Aug 5, 2006 6:22pm
FF.........

This does not make sense...........I use CSI and get my data NO LATER than 5:30 ET, usually by 5:15 ET. . Still, I do not see how your system will work because pricing at 5 PM ET is usually always very close to the p/m on all 10 pairs I trade with VP. When it isn't, it can be very tricky to accurately predict which direction the price will go first before hitting the p/m, if at all. I remember that at the beginning of our forward test project with VP (back in Feb, I think), someone posted information about trading toward the p/m 12 hours after the data was released and that this strategy produced very good results. Anyone here remember that?

BB


Quoting floridaforex
That is one of the main problems with my system. Because I get the data so late I cannot usually trade it, so it's only good in theory.

I haven't been able to get my data before 6:30. Yesterday I tried downloading it at 6:25 and it wasn't updated yet. I tried again 5 minutes later and it worked. I'd love to get my data at 5:15.

I just got a hold of hourly intraday data going back to Jan05. I'll let you know what I find out.

Do you get your data at 5 EST or central time? I'm never able to get it that early and I use CSI.
  • #618
  • Aug 5, 2006 8:40pm Aug 5, 2006 8:40pm
I think there is a setting in the CSI ezdownloader for what server it goes through. Maybe yours is updated before mine. When I signed up it even told me that the updates wouldn't be out until 6:30 ET. This is really weird, I don't see why they would make it so some get it early and others late.

What are the server settings for your CSI download program?
  • #619
  • Aug 5, 2006 8:56pm Aug 5, 2006 8:56pm
Quoting floridaforex
I think there is a setting in the CSI ezdownloader for what server it goes through. Maybe yours is updated before mine. When I signed up it even told me that the updates wouldn't be out until 6:30 ET. This is really weird, I don't see why they would make it so some get it early and others late.

What are the server settings for your CSI download program?
At the top of CSI's Unfair Advantage screen, the fourth heading is Database.
A left click produces "View Posting Status". This will show precisely when your data is available by turning from red to blue in the Forex (third) box from the left. It carries "4.50" but my data is almost always available by 17.15 (5.15pm) EST.

BTW, where do you find the server settings?
  • #620
  • Aug 5, 2006 9:03pm Aug 5, 2006 9:03pm
Got it. My primary server is ua2.csidata.com
1 trader viewing now