USD/JPY Discussion | Page 215 | Forex Factory

Options

Go to
Permalinks
Print Thread

Related Instruments

USD/JPY

Similar Threads

USD/JPY Discussion 12 replies

NZD/JPY Discussion 12 replies

long eur/jpy, gbp/jpy, usd/jpy 11 replies

EUR/USD Bollinger Band Discussion 3 replies

Suidster's GBP/JPY Discussion 19 replies

Attachments: USD/JPY Discussion
Exit Attachments

USD/JPY Discussion

  • #4,281
  • Aug 27, 2007 5:17pm Aug 27, 2007 5:17pm
Quoting Trader KGB
I've backtested the pivots on U/J and found no statistically significant evidence that if one pivot level is broken, the next pivot must be hit. Even though recent history might suggest otherwise, I could show you 100 intraday charts where the first S/R pivot level was broken and nothing came of it.

In this case, we haven't broken the S1 on the 5-minute. Visually, pivot touches that lead to reversals appear to occur more often than would otherwise be deduced as random behavior. I haven't yet been able to program a backtest simulation for these rules, so I'm not comfortable to state anything conclusively.

Looking at the current chart, I'd say the S2 is possible. We weren't able to eclipse last Wed's high, and we're only a few pips above last Thur's low. If this cycle of lower highs and lower lows continues, then probability will favor the bearish side.
What else have you learned from backtesting?
  • #4,282
  • Aug 27, 2007 5:25pm Aug 27, 2007 5:25pm
Quoting Trader KGB
I've backtested the pivots on U/J and found no statistically significant evidence that if one pivot level is broken, the next pivot must be hit. Even though recent history might suggest otherwise, I could show you 100 intraday charts where the first S/R pivot level was broken and nothing came of it.

In this case, we haven't broken the S1 on the 5-minute. Visually, pivot touches that lead to reversals appear to occur more often than would otherwise be deduced as random behavior. I haven't yet been able to program a backtest simulation for these rules, so I'm not comfortable to state anything conclusively.

Looking at the current chart, I'd say the S2 is possible. We weren't able to eclipse last Wed's high, and we're only a few pips above last Thur's low. If this cycle of lower highs and lower lows continues, then probability will favor the bearish side.
I'm not sure if the S/R levels are the only pivots that matter in those cases. there are mid's, weekly's, 50%'ers etc.
i don't think you could back test something like that.
  • #4,283
  • Aug 27, 2007 6:26pm Aug 27, 2007 6:26pm
Quoting kreeper_6
I'm not sure if the S/R levels are the only pivots that matter in those cases. there are mid's, weekly's, 50%'ers etc.
i don't think you could back test something like that.
You can. It just takes a good deal of work.. I could do it in MT4 today (already have the mid's, weekly's, monthly's coded in), the problem is, I don't have much trust in the datafeed quality of the MT4 brokers, especially on a historical intraday level. The culprits are too many to count (Sunday candles, non-standard daily closes at whatever timezone the broker chooses, etc).

Admittedly, the first backtest based upon the S/R levels was done in MT4, so therefore, take my results with a grain of salt. It has been relatively reliable in calculating the daily pivots (with the only mt4 broker that uses a proper 17:00 close), so I felt it capable of running a rudimentary backtest. Anything involving mid's, central's, weekly's, monthy's will require more work, and such time investment should be done on a more reliable datafeed/platform (probably Amibroker for the platform, still investigating the data provider).
  • #4,284
  • Aug 27, 2007 6:39pm Aug 27, 2007 6:39pm
Quoting trader07
What else have you learned from backtesting?
Just what was earlier stated. It was just a quick test of the Dove Theory, that a pivot break in U/J leads price to the next pivot in succession by the 17:00 close.
  • #4,285
  • Aug 27, 2007 6:56pm Aug 27, 2007 6:56pm
Could we be setting up to trade between the S1 and the Daily???
creating wealth one pip at a time.....
  • #4,286
  • Aug 27, 2007 7:08pm Aug 27, 2007 7:08pm
Quoting Trader KGB
You can. It just takes a good deal of work.. I could do it in MT4 today (already have the mid's, weekly's, monthly's coded in), the problem is, I don't have much trust in the datafeed quality of the MT4 brokers, especially on a historical intraday level. The culprits are too many to count (Sunday candles, non-standard daily closes at whatever timezone the broker chooses, etc).

Admittedly, the first backtest based upon the S/R levels was done in MT4, so therefore, take my results with a grain of salt. It has been relatively reliable in calculating the daily pivots (with the only mt4 broker that uses a proper 17:00 close), so I felt it capable of running a rudimentary backtest. Anything involving mid's, central's, weekly's, monthy's will require more work, and such time investment should be done on a more reliable datafeed/platform (probably Amibroker for the platform, still investigating the data provider).
do you have 90% quality, if not I can send you a link on how to do so.

alpari data is the data you want for setting up backtesting.
  • #4,287
  • Aug 27, 2007 7:10pm Aug 27, 2007 7:10pm
Quoting wheelsut
Could we be setting up to trade between the S1 and the Daily???
uhhhhh, I've been meaning to ask about your avatar.

is it yours?
  • #4,288
  • Aug 27, 2007 7:22pm Aug 27, 2007 7:22pm
Quoting kreeper_6
uhhhhh, I've been meaning to ask about your avatar.

is it yours?
Yes it is my avatar LMAO, I am a baby to this whole thing, thought it was apporpriate but thanks for asking!!!
creating wealth one pip at a time.....
  • #4,289
  • Aug 27, 2007 7:37pm Aug 27, 2007 7:37pm
Quoting kreeper_6
uhhhhh, I've been meaning to ask about your avatar.

is it yours?
LOL!
  • #4,290
  • Aug 27, 2007 7:53pm Aug 27, 2007 7:53pm
  • #4,291
  • Aug 27, 2007 8:21pm Aug 27, 2007 8:21pm
hey guys.. do we have a new daily and pivots yet?
  • #4,292
  • Aug 27, 2007 8:24pm Aug 27, 2007 8:24pm
Quoting invicta79
hey guys.. do we have a new daily and pivots yet?
Top of last page.
  • #4,293
  • Aug 27, 2007 8:44pm Aug 27, 2007 8:44pm
Quoting kreeper_6
do you have 90% quality, if not I can send you a link on how to do so.

alpari data is the data you want for setting up backtesting.
Quality varies, please send the link.

Alpari uses an 18:00 close... All pivot calculations would have to be done with a 1-hour time offset....
  • #4,294
  • Aug 27, 2007 10:14pm Aug 27, 2007 10:14pm
Price is confined in between the 50 % fib (yellow) and the 50% fib (blue)/weekly pivot point (purple line). The yellow fibs are drawn on the high and low of last week.

The attached charts are on a 2HR timeframe with the latter being a closeup view of the first.
Attached Image(s) (click to enlarge)
Click to Enlarge Name: 2HR DOLLAR YEN.GIF Size: 33 KB
Click to Enlarge Name: 2HR DOLLAR YEN CLOSEUP.GIF Size: 29 KB
  • #4,295
  • Aug 27, 2007 10:16pm Aug 27, 2007 10:16pm
Quoting Trader KGB
Quality varies, please send the link.

Alpari uses an 18:00 close... All pivot calculations would have to be done with a 1-hour time offset....
Here is the pdf how to. You have to use a mt4 install that is offline though. you can install another copy of mt4 just by installing into a different directory.

Enjoy...
Attached File(s)
File Type: pdf how-get-modelling-quality-90-when-testing-ea-mt4.pdf   331 KB | 318 downloads
  • #4,296
  • Aug 27, 2007 10:32pm Aug 27, 2007 10:32pm
Quoting kreeper_6
Here is the pdf how to. You have to use a mt4 install that is offline though. you can install another copy of mt4 just by installing into a different directory.

Enjoy...
Thanks. Since it's required to run in offline mode, I'm hoping there may be a way to overcome the server close of 18:00..
  • #4,297
  • Aug 27, 2007 10:44pm Aug 27, 2007 10:44pm
i shorted at 115.64... made 10 pips so far... would a 115.25 limit be reasonable?
  • #4,298
  • Aug 28, 2007 2:00am Aug 28, 2007 2:00am
Guys,

This is not directed at any specific poster but I want to stress that it is not in anyones best interest to keep asking questions like, 'Should I buy here' and 'Is my long position safe here or shall I increase my stop'. It shows no understanding of what has been posted on this thread.

This thread started off as a general discussion about USDJPY, and then we were fortunate enough to have Dove enlighten us with his trading perspectives and add some greater direction to this thread.

Asking questions is fine, we alll need to do this to learn, but please structure them in a way so that those who may provide the most helpful answers know that you will be listening and trying to learn rather than looking for trade recommendations.

For example, 'Price broke down through S1 and then had a small retrace back to S1 so I enetred a short position as the trend appeared down. I have a target of S2 and a protective stop 15 pips above S1. Is this a good trade'.

Good trading all.

G-Man
  • #4,299
  • Aug 28, 2007 2:13am Aug 28, 2007 2:13am
Quoting invicta79
i shorted at 115.64... made 10 pips so far... would a 115.25 limit be reasonable?
good call.
although last 3 hours are bullish i believe the pair will be bearish.
115.37 is my first target
if closed below it than I will aim 114.72
if closed above 115.70 than i am quiting my position
give me my pips back...
  • #4,300
  • Aug 28, 2007 2:30am Aug 28, 2007 2:30am
Quoting Lincoln_10
good call.
although last 3 hours are bullish i believe the pair will be bearish.
115.37 is my first target
if closed below it than I will aim 114.72
if closed above 115.70 than i am quiting my position
Hey Im with you on that trade too
entered at 115.56 with a PT at 115.30 (support level) while SL is at 115.80 (resistance level)
"Don't work for your money, but let money work for you."
27 traders viewing now