XAO - Opening 10 minutes as prediction for the day?

[Gringotts Bank]

Unless the futures are providing a very strong morning lead (say +/-50), then they seem to have very little predictive power for XAO on the same day.

Similarly, the DOW has almost no predictive power. Sometimes the two indices do come strongly into synch, but it's not reliable.


I wonder how well the first 10 minutes of trade predicts whether the Ords finishes above or below the line. Today is another one of those +/-/+/- days. But as a general rule, I would have thought the opening swing would be reasonably predictive of a + or - close for the index. What do you think?

Anyone with XAO tick data test this?

buy = 10:10 am.
sell = close

Thanks.

 

[sinner]

Unless the futures are providing a very strong morning lead (say +/-50), then they seem to have very little predictive power for XAO on the same day.

Similarly, the DOW has almost no predictive power. Sometimes the two indices do come strongly into synch, but it's not reliable.


I wonder how well the first 10 minutes of trade predicts whether the Ords finishes above or below the line. Today is another one of those +/-/+/- days. But as a general rule, I would have thought the opening swing would be reasonably predictive of a + or - close for the index. What do you think?

Anyone with XAO tick data test this?

buy = 10:10 am.
sell = close

Thanks.

Do you mean, as in an Open Range Breakout style of trade?

My own testing shows this only seems to work on various instruments, rather than all of them. Make sure you are considering some instrument you can actually trade, rather than an untradeable calculated index, the opening swing on XJO is probably represented as a gap on SPI futures (or STW).

For example GBPUSD (London Open) and the Hang Seng (HK Open) both seem to provide a positive expectancy for 30min ORBs.

I doubt the XJO would, even though I haven't tested it.

There is a paper:
Trading System Development: Trading the Opening Range Breakouts
ME Cekirdekci - 2010

That goes into it.

 

[Gringotts Bank]

Thanks sinner.

I guess I just just want to know, "if the XAO is green at say 10:10am, will it close the day green?" then take a position in stock xyz.

I very roughly tested it using EOD data of XAO, and 2/3 of the time this above query is true, (maybe more than 2/3). I'd just like some specific figures. Might need to buy myself some tick data.

Thanks for that paper. Looks good.

 

[tech/a]

GG

Its common practice to cover shorts or longs on open.
As such its normally wise to fade the opening 10 mins or so.

What your seeing is normal.

 

[Gringotts Bank]

So a fund will cover shorts or sell longs at open, creating volatility in the opening minutes, sure.

Why would that mean you should fade the open after 10 minutes?

 

[sinner]

Thanks sinner.

I guess I just just want to know, "if the XAO is green at say 10:10am, will it close the day green?" then take a position in stock xyz.

I very roughly tested it using EOD data of XAO, and 2/3 of the time this above query is true, (maybe more than 2/3). I'd just like some specific figures. Might need to buy myself some tick data.

Thanks for that paper. Looks good.

Ok so eally what you are saying here isn't ORB, it is IF Close > Open for the first 10min bar, then bias accordingly.

Hmmm again, I think you are not using the right instrument by checking the XAO. Doesn't the All Ords officially price open where it priced closed the prior day? It's not a real instrument, and your bias will not match accordingly. Take a look at the STW ETF for example. I think if you test using the XAO then your bias will pretty much look like that chart.

From an intraday perspective, I think it would be a wiser to consider a "WHILE" conditional, rather than "IF" conditional.

i.e.

"WHILE XAO > open price, long xyz stock" (might be better to use STW than XAO, honestly)

So a fund will cover shorts or sell longs at open, creating volatility in the opening minutes, sure.

Why would that mean you should fade the open after 10 minutes?

Realistically, it's different in bear markets from bull markets. What you describe as the Close > Open in XAO, is the opening gap in tradeable instruments really I think. This gap does have bias, but it's pretty dependent on other factors.

But tech is generally right that the expectancy is to fade (imho).

 

[Gringotts Bank]

Yeh STW probably more appropriate, or even BHP.

I see what you're saying tech - fade the upwards gap.

 

[sinner]

Yeh STW probably more appropriate, or even BHP.

I see what you're saying tech - fade the upwards gap.

Now that we know what you are actually trying to do/use,

here is some appropriate stuff from TH

http://tremblinghandtrader.typepad.com/trembling_hand_trader/2007/07/bear-gaps.html
http://tremblinghandtrader.typepad....2010/02/the-problem-with-the-asx-indexes.html

Both posts have done a lot of the hard work for you.

 

[Gringotts Bank]

Yeh that wrecks that method!

Thanks for ideas.

 

[Wysiwyg]

http://tremblinghandtrader.typepad.com/trembling_hand_trader/2007/07/bear-gaps.html
Both posts have done a lot of the hard work for you.

The test for fading the SPI opening gap down was over one time period and we know that one time period tests are of no statistical value for trading. I have little evidence of definitive price movements assumed from back testing. He types this about the one period test ....

Only a slight advantage getting long on gaps down yet still enough for me to think down open = go short is not the way you should play.

Yes was a surprise that it was so close to 50/50 on drops.

Posted by: Trembling Hand Trader | July 25, 2007 at 01:26 AM

 

[sinner]

The test for fading the SPI opening gap down was over one time period and we know that one time period tests are of no statistical value for trading. I have little evidence of definitive price movements assumed from back testing. He types this about the one period test ....

Doctor Wysiwyg at it again, with his expert opinion...



Just a few questions for the Doctor,

Was any information presented as having statistical value for trading? (or was it presented in the interests of discussion and education)
Does your "evidence" actually have anything to do with the topic at hand, or is this like the last thread with backtesting discussion that you decided to derail? Your strange obsession with convincing people that backtesting doesn't work, or whatever it is you're trying to convince people of?

I am sure GB is quite capable of reading the links and discerning what is written plain as day on the website without needing you to point it out for him. I certainly don't need it from you.

Of course I find it slightly amusing you didn't quote this

This is only a simple example that really isn't a trading system but its one reason why you need to be trading with what is ahead rather than whats happened in the rearview mirror.

 

[Wysiwyg]

Doctor Wysiwyg at it again, with his expert opinion...

No doubt about it, you're good, but I am the expert here ol' chap. However, in acknowledgment of your intellectual prowess, could you let us know when this bear market cycle bottoms please. I wish to profit from the next run to an all time high.