Australian (ASX) Stock Market Forum

What is the definition of a Big White candlestick?

Joined
18 February 2010
Posts
65
Reaction score
1
I had some very good results backtesting in Bullcharts and Tradesim a system using Big White candlesticks. I’d like to duplicate it in Excel. Bullcharts formula must be proprietary, because I couldn’t get into the guts of it.

I couldn’t find a formula or definition of a Big White [Tower/Real Body] anywhere on the net. Everyone talks vaguely about “tall” “long” and “big” candlesticks but no-one defines precisely when is a Big White not a Big White.

I ended up defining it as:

If today’s real body length is 2 times the average real body over the last 100 periods. [arbitrary parameters of course.]

Most times this seems to match proprietary software’s candlesticks marked Big White, but not always.

Just wondering if someone has a better formula/definition.

Thank you for your help.
 
Joined
30 June 2007
Posts
6,175
Reaction score
619
I had some very good results backtesting in Bullcharts and Tradesim a system using Big White candlesticks. I’d like to duplicate it in Excel. Bullcharts formula must be proprietary, because I couldn’t get into the guts of it.

I couldn’t find a formula or definition of a Big White [Tower/Real Body] anywhere on the net. Everyone talks vaguely about “tall” “long” and “big” candlesticks but no-one defines precisely when is a Big White not a Big White.

I ended up defining it as:

If today’s real body length is 2 times the average real body over the last 100 periods. [arbitrary parameters of course.]

Most times this seems to match proprietary software’s candlesticks marked Big White, but not always.

Just wondering if someone has a better formula/definition.

Thank you for your help.

For such a test you are going to want to optimize both parameters, so you do that over a large range.

eg. length = 1.2 - 5 and periods = 10-300
 

CanOz

Home runs feel good, but base hits pay bills!
Joined
11 July 2006
Posts
11,534
Reaction score
495
I had some very good results backtesting in Bullcharts and Tradesim a system using Big White candlesticks. I’d like to duplicate it in Excel. Bullcharts formula must be proprietary, because I couldn’t get into the guts of it.

I couldn’t find a formula or definition of a Big White [Tower/Real Body] anywhere on the net. Everyone talks vaguely about “tall” “long” and “big” candlesticks but no-one defines precisely when is a Big White not a Big White.

I ended up defining it as:

If today’s real body length is 2 times the average real body over the last 100 periods. [arbitrary parameters of course.]

Most times this seems to match proprietary software’s candlesticks marked Big White, but not always.

Just wondering if someone has a better formula/definition.

Thank you for your help.

I think you'd be on the right track, its obviously some kind of ATR system....I've seen one in a book i have somewhere that's like 200% greater than the last x bars...Maybe you could optimize a % over a small amount of data to see a typical range that works ok?

CanOz
 
Joined
18 February 2010
Posts
65
Reaction score
1
Thank you Gringotts and Canoz. I take your point: vary my parameter inputs and maybe come up with something even better than the software proprietary definition, which in itself is only arbitrary anyway. A job for the weekend.

Will report back.
 
Joined
18 February 2010
Posts
65
Reaction score
1
Thank you Gringotts and Canoz. I take your point: vary my parameter inputs and maybe come up with something even better than the software proprietary definition, which in itself is only arbitrary anyway. A job for the weekend.

Will report back.

My conclusions were...the blinking obvious really!

The rarer/more exceptional you make the Big White by increasing the number of days you are averaging back the real body, and, more importantly, increasing the size of the Big White..from 1.2 to 3 times the average real body over 100 to 300 days...the better the results. But being rarer that means the less trading and the longer in between Big Whites..which may not be a bad thing, of course, in order to try to improve your odds of success.

I settled in the end on 200 period lookback for the average real body size and 1.75 * the average real body size to define the Big White.

In addition, a shaven bottom candlestick or very small lower wick (<5% of that day’s real body) on entry improves results.

The Profit Index and Win/Loss ratio are exceptionally good if you can jump aboard one of these halfway between the open and the close. Profits deteriorate if you wait till the close. Now there’s the Catch 22. How do you know it’s going to be a solid Big White half way through the trading day ..it might turn into a Long Upper Shadow? I suppose you could jump aboard on an intraday pullback or at 1.5 *real body if other indicators support the strength being there.

Eg

Success: SXL enter 9 May @ 153.5 (open+1.5*its average real body) Exit 14 May @160 falling momentum.
Failure: CFX enter 24 April @ 222 (open+1.5*its average real body) Exit 26 April @221 falling momentum.

Then again re CFX you could make a rule that, if the candle doesn’t end the day as a Big White (i.e with shaven head or a very small upper wick) and the profit takers have already crept in (see the sizable upper shadow), then it’s not behaving the way you want it to, so grab your modest profit if you can just before close @223 and run..there’ll be others.

However, the results so far are an encouragement to research further.
 
Joined
30 June 2007
Posts
6,175
Reaction score
619
Now there’s the Catch 22. How do you know it’s going to be a solid Big White half way through the trading day ..it might turn into a Long Upper Shadow? I suppose you could jump aboard on an intraday pullback or at 1.5 *real body if other indicators support the strength being there.

It won't work. I've tried this sort of thing myself.

If you are really keen on this idea and are willing to spend a lot of money to use Iguana Spark data, there is a way to do it and profit, but you have to be watching live. PM me if interested.
 

tech/a

No Ordinary Duck
Joined
14 October 2004
Posts
19,976
Reaction score
5,047
I think as a tool for practical application in a technical trading method
Your missing an important ingredient.
VOLUME
The other important criteria has been addressed MAGNITUDE OF RANGE.
Metastock has a charting tool which charts a candle relative to Volume.
Think it's called Eqivolume.

I'm sure when you investigate Big White Candles will signal a different
Outcome if coupled with/Very high volume/ High Volume/ Average Volume/ Low volume
Very low volume.

I've done a lot of work with wide range bars ( Big White Candles )
 
Joined
18 February 2010
Posts
65
Reaction score
1
I think as a tool for practical application in a technical trading method
Your missing an important ingredient.
VOLUME
The other important criteria has been addressed MAGNITUDE OF RANGE.
Metastock has a charting tool which charts a candle relative to Volume.
Think it's called Eqivolume.

I'm sure when you investigate Big White Candles will signal a different
Outcome if coupled with/Very high volume/ High Volume/ Average Volume/ Low volume
Very low volume.

I've done a lot of work with wide range bars ( Big White Candles )

Thank you Tech/a. I have seen the Equivolume feature on BC, but never got around to using it..just got too used to regular candlestick patterns. But I’ll give them a go.

I will add volume to the picture too.

BTW, if the candlestick prior to the BigWhite is a doji, harami, morning star it improves results too...reflecting punters’ indecision turning to bullishness (or fake bullishness if there are sharks in the water!). Going to test other situations too. Again, makes the occurrences rarer, but that may be no bad thing if they are more successful.

Thank you most kindly.
 

tech/a

No Ordinary Duck
Joined
14 October 2004
Posts
19,976
Reaction score
5,047
Thank you Tech/a. I have seen the Equivolume feature on BC, but never got around to using it..just got too used to regular candlestick patterns. But I’ll give them a go.

I will add volume to the picture too.

BTW, if the candlestick prior to the BigWhite is a doji, harami, morning star it improves results too...reflecting punters’ indecision turning to bullishness (or fake bullishness if there are sharks in the water!). Going to test other situations too. Again, makes the occurrences rarer, but that may be no bad thing if they are more successful.

Thank you most kindly.

You bring up a strong point
Finding a likely setup for a big White Candle
Surely that's preferable to what happens after it
As this would become the impulse or control bar
 
Joined
18 February 2010
Posts
65
Reaction score
1
It won't work. I've tried this sort of thing myself.

If you are really keen on this idea and are willing to spend a lot of money to use Iguana Spark data, there is a way to do it and profit, but you have to be watching live. PM me if interested.

Thanks for your help, GB and Tech/A. Tried to PM thanks for info to you GB, but bounced back...your PM box full.

Besides the setup candlesticks mentioned above, I also improved results with gaps, especially...

in shorting, where today opened <open and <close of the previous day (body gap), but either was still not less than the previous low, or tried to get back up into that previous low area and failed i.e....and high>=hist(low,1)

Back to reading the tea leaves.
 

CanOz

Home runs feel good, but base hits pay bills!
Joined
11 July 2006
Posts
11,534
Reaction score
495
How many of these types of BWC's occur after an inside day?

CanOz
 
Joined
26 October 2008
Posts
2,931
Reaction score
4
Thinking about it in terms of candlesticks is always going to confuse the issue.

I much prefer to think in terms of Crabel, i.e. NR(2,3,4), NR(5,6,7), WR(2,3,4), WR(5,6,7). Combine this with Internal Bar Strength and you have all the quantitative information you need to build these kinds of systems.

Historically, Narrow Ranges precede Wide Ranges, in all markets.
 
Joined
30 June 2007
Posts
6,175
Reaction score
619
Thinking about it in terms of candlesticks is always going to confuse the issue.

I much prefer to think in terms of Crabel, i.e. NR(2,3,4), NR(5,6,7), WR(2,3,4), WR(5,6,7). Combine this with Internal Bar Strength and you have all the quantitative information you need to build these kinds of systems.

Historically, Narrow Ranges precede Wide Ranges, in all markets.

What is "internal bar strength" please?
 
Joined
30 June 2007
Posts
6,175
Reaction score
619
What is "internal bar strength" please?

I was being lazy!

The location of the closing price within the day's range is a surprisingly powerful predictor of next-day returns for equity indices. The closing price in relation to the day's range:Internal Bar Strength is simply calculated as such:

((close-low)/(high-low))*100

It takes values between 0 and 100 and simply indicates at which point along the day's range the closing price is located. In this post I will take a look not only at returns forecasting, but also how to use this value in conjunction with other indicators. You may be skeptical about the value of something so extremely simplistic, but I think you'll be pleasantly surprised.

Excellent research related to IBS you can find here:
http://qusma.com/2012/11/06/closing...e-days-range-and-equity-index-mean-reversion/

Read more: http://www.quantshare.com/item-1242-ibs-internal-bar-strength#ixzz2VDe9UUNc
Follow us: @quantshare on Twitter
 
Joined
18 February 2010
Posts
65
Reaction score
1
How many of these types of BWC's occur after an inside day?

CanOz

On the SHORT side.

Using Harami as the definition of an inside day, since it’s already programmed into BC.

Using the XJO over approx last 3 years, 205 instances ...traded as is, it may be a losing system unless you trade it very short term entering at a reasonable price (intraday pullback?).

Throw in other factors such as the stock already trending down or today marks a trend reversal or a trendbreak plus, as Tech/a said, unusual Volume ...the # of instances falls and the system becomes more profitable.

Thanks for the info, Sinner.

I had never heard of Toby Crabel, but Googling certainly throws up some interesting ideas.

It’s not only back to the tea leaves for me, it’s back to the drawing board and coding.
 
Top