| Author |
Message |
     button_presser
Member
Username: button_presser
Post Number: 1
Registered: 11-2003Rating: N/A
Votes: 0
|
| | Wednesday, February 04, 2004 - 11:14 am: |  
|
Just wondering if the IC team had in mind a 'build-your-own indicator' module. I would love to be able to personalise my own indicators as I have researched some interesting adaptations of the more common indicators available.
On the same topic, the filtering section would benefit from a 'build-your-own filter' module.
I am aware that Metastock has these features but have not yet come across any comments/opinions to do with them specifically.
Is Metastock worth a try? Any comments?
Thanks
|
colin_twiggs
Member
Username: colin_twiggs
Post Number: 987
Registered: 09-2002Rating: N/A
Votes: 0
|
| | Wednesday, February 04, 2004 - 12:01 pm: |
|
BP
We have discussed the idea, but there is enough on our To Do list to keep us busy until 2005. Perhaps then we can look at it more seriously.
Colin
INCREDIBLE OFFER - only $12 USD per month
30 day FREE trial -
Download Here
|
|
|
|
|
williamat
Member
Username: williamat
Post Number: 194
Registered: 09-2002Rating: N/A
Votes: 0
|
| | Thursday, February 05, 2004 - 08:21 am: |
|
BP,
from discussions and reading Forum, it would seem that Amibroker (introduced by Bundy) and Market analyst which has been mentioned several times would both be worthy of your investigation. You can find them on the net by searching for the names, and they each have free trial periods.
Bill.
The difference between intelligence is this- intelligence will make you a good living. Charles C Kettering.
|
mdees1
Member
Username: mdees1
Post Number: 6
Registered: 10-2006Rating: N/A
Votes: 0
|
| | Monday, March 05, 2007 - 03:21 pm: |
|
Just wondering if any progress was made regarding a build your own indicator module???
|
colin_twiggs
Member
Username: colin_twiggs
Post Number: 2830
Registered: 09-2002Rating: N/A
Votes: 0
|
| | Thursday, March 08, 2007 - 02:13 pm: |
|
No progress as yet. Our current development priorities are:
(1) Fibonacci Tools
(2) Stock Screener (upgrade)
(3) Indicators (upgrade)
(Don't worry Dug. Haven't forgotten VWAP. We are working on getting the data in place.)
Regards,
Colin
|
hilarius
Member
Username: hilarius
Post Number: 2119
Registered: 04-2004
Rating: N/A
Votes: 0
|
| | Thursday, March 08, 2007 - 04:11 pm: |
|
Hello Colin
It would be greatly appreciated if it could be possible to adjust the look back period for Volume Spike scanning ... eg 5 and 10 day look back options
Hilarius
I come in peace to share my thoughts and to shine my candle light on possible long term opportunities
|
colin_twiggs
Member
Username: colin_twiggs
Post Number: 2832
Registered: 09-2002Rating: N/A
Votes: 0
|
| | Friday, March 09, 2007 - 09:11 am: |
|
Roger,
Volume Spikes do have a look-back period. You can scan for volume spikes that occurred within the last x number of days.
Regards,
Colin
|
hilarius
Member
Username: hilarius
Post Number: 2121
Registered: 04-2004
Rating: N/A
Votes: 0
|
| | Friday, March 09, 2007 - 09:58 am: |
|
Thanks for that Colin
I should have been much clearer in my request
What I was looking for was the ability to vary the length of the exponential moving average volume to as little as 10 or even 5 days
I just raised it because I saw you are in the process of upgrading the scanning tools
Apologies for my lack of clarity
Hilarius
PS I am in a group of users scanning for volume spikes and we would value the shorter moving average period so that our searches can be standardised ... and made less dependent on cumbersome downloading of data from brokers and Excel querying. We are keen to optimise our use of IC scans.
I come in peace to share my thoughts and to shine my candle light on possible long term opportunities
|
colin_twiggs
Member
Username: colin_twiggs
Post Number: 2833
Registered: 09-2002Rating: N/A
Votes: 0
|
| | Friday, March 09, 2007 - 10:12 am: |
|
Hilarius,
The 10-day or 5-day MA will be more volatile and therefore give a more uneven set of results: the same volume may be classed as a spike one day and not the next. Not what I think you are trying to achieve.
How do you want to use this filter?
Regards, Colin
|
holycow
Member
Username: holycow
Post Number: 2778
Registered: 08-2004
Rating: N/A
Votes: 0
|
| | Friday, March 09, 2007 - 11:10 am: |
|
I found volume spike to have a long lasting residual effect, especially in the early or midterm phase of a rallying market. It is an early alert that something is happening (or has happened) with the stock - explanation such as substantial ownership changing hand is one possibility, takeover ploy, M&A activity spilled into open market, etc are others.
To be able to back scan for vol spikes I guess could be useful for someone like me who subscribes to this "theory". The alternative is to keep a journal of unusual vol spikes for on going monitoring.
HC
"... I believe in Santa!"
|
hilarius
Member
Username: hilarius
Post Number: 2122
Registered: 04-2004
Rating: N/A
Votes: 0
|
| | Saturday, March 10, 2007 - 10:52 am: |
|
Colin
Thanks for the comment on the request for other options for the length of the volume moving average
A consensus has now developed in the user group to continue using your 50 day moving average
(I retain a personal desire to look at a 20 day moving average approximating 1 month of trading ... but will study how many extra or fewer results it gives before formally requesting it.)
However we now have a query about the way the 50 day moving average operates
The example I have is BOQ which appears in your 1.5 times listing, but on my simple average calculation is a 2.2 times stock
Would the difference between ema and sma really make that much difference? I guess it could but would value it greatly if you could check whether we are using the same volume data as IC
I am attaching a file of my BOQ 50 day sma calculation
With Best Wishes
Hilarius
PS If anyone could attach the ema calculation to the file and reload it that would be very interesting information also ... my skills stop short of the ema formula
I come in peace to share my thoughts and to shine my candle light on possible long term opportunities
|
colin_twiggs
Member
Username: colin_twiggs
Post Number: 2834
Registered: 09-2002Rating: N/A
Votes: 0
|
| | Saturday, March 10, 2007 - 01:13 pm: |
|
Roger,
Your Volume data appears lower than IC. Thursday your file shows volume of 249,657 while IC shows 300,390. It could be that you are picking up the 4pm closing volume whereas IC will include volume on after market trading.
Compare the attached file to your results on IC.
Regards,
Colin
|
hilarius
Member
Username: hilarius
Post Number: 2123
Registered: 04-2004
Rating: N/A
Votes: 0
|
| | Saturday, March 10, 2007 - 05:09 pm: |
|
Colin
Thanks for the EMA calculation ... greatly appreciated
I've just downloaded the volumes for BOQ for 51 days from my online broker and only one day is different to my above Excel file, namely 26/2/07 which has dropped from 272,110 to 169,792
Other days do not seem to have changed from the figures which I download around 4.45 pm each day giving me the impression they do not update to the final figures which you have
I would like to view the final figures for BOQ for the period 27/12/06 to 9/3/07 inclusive to see how they compare to the 4.45 pm figures ... it seems that the later figures do give rise to higher average volumes and thus smaller ratios of spike volume to average volume
The difference seems to be sufficient to have pushed BOQ down from the 2.0 times group to the 1.5 times group
My ongoing problem is how to access the final volumes rather than the interim 4.45 pm ones, for all stocks, since I wish to build a database of volumes to use offline.
Any suggestions would be appreciated
With Best Wishes
Hilarius
I come in peace to share my thoughts and to shine my candle light on possible long term opportunities
|
colin_twiggs
Member
Username: colin_twiggs
Post Number: 2836
Registered: 09-2002Rating: N/A
Votes: 0
|
| | Monday, March 12, 2007 - 11:51 am: |
|
Roger,
How does the data compare to Yahoo?
Regards,
Colin
|
hilarius
Member
Username: hilarius
Post Number: 2126
Registered: 04-2004
Rating: N/A
Votes: 0
|
| | Friday, March 16, 2007 - 03:40 am: |
|
Good Morning Colin
Herewith a file showing calculations of 50 day EMA of Volume for BOQ as at Friday 9th March using data from :-
(1) Online Broker
(2) Yahoo
(3) IC
In all 3 cases I get a result in which the actual volume is greater than 2 times the EMA, for Friday 9th March
However, I believe that in the IC Volume Scan for Friday 9th BOQ appeared only in the 1.5 times Scan
At this stage I can't recall the time of day I did the Friday scan, but it was after the 4pm charts had appeared for Friday
When I do the scan now with the significant volumes in the last four days BOQ has moved into your 2 times scan
However, the 2 times trigger was needed after last Friday's scan (9th March) for our current system to activate a BUY, while a 1.5 times result is not considered
As you can see a BUY signal after last Friday's trading would have given a very attractive entry
I am wondering whether you have any means of reconstructing the scan for last Friday, the 9th, to determine whether or not it was then truly a 1.5 times situation or a 2 times situation, per your methodology?
It makes a big difference to our system and I just hope I am not mistaken in recalling the 1.5 times result for BOQ when I did the scan after last Friday's trading
I really hope this is not a "wild goose chase" for you but the 1.5 times result is clearly in my mind and I did the scan a couple of times to confirm it. Unfortunately I did not keep a copy of the scan result in a watchlist.
Whether there is a technical issue causing the 1.5 times result versus my calculated 2, or if my calculation is wrong, I would like to resolve the problem
I am reasonably sure of all the facts as given above but apologise in advance if I am shown to be suffering from a brain malfunction ... which is always a possibility to consider
Sadly I was not under the influence of a cellared red wine or two at the time I did the maths ... that might have helped 
With Best Wishes
Hilarius
I come in peace to share my thoughts and to shine my candle light on possible long term opportunities
|
hilarius
Member
Username: hilarius
Post Number: 2127
Registered: 04-2004
Rating: N/A
Votes: 0
|
| | Friday, March 16, 2007 - 03:46 am: |
|
Chart of BOQ :-
I come in peace to share my thoughts and to shine my candle light on possible long term opportunities
|
ohkoolnutz
Member
Username: ohkoolnutz
Post Number: 544
Registered: 10-2005
Rating: N/A
Votes: 0
|
| | Friday, March 16, 2007 - 10:27 pm: |
|
It's a good example how scans lag the action in (too) many cases. I bought BOQ on March 8 at $15.59 which was two days ahead of a possible scan result buy.
---
ohk
Lies, Damn Lies and Technical Analysis
|
hilarius
Member
Username: hilarius
Post Number: 2128
Registered: 04-2004
Rating: N/A
Votes: 0
|
| | Friday, March 16, 2007 - 10:50 pm: |
|
OHK
The data I have provided will hopefully enable Colin to determine whether there was a calculation error in the Friday 9th scan ... or whether my memory is at fault in recalling there was a 1.5 times volume spike ... when for me a 2 times volume spike would have been much more conclusive
So at this stage the maths need to be resolved ... a 2 times scan result on Friday 9th would have been just fine, from my perspective.
With Best Wishes
I come in peace to share my thoughts and to shine my candle light on possible long term opportunities
|
msparks
Member
Username: msparks
Post Number: 819
Registered: 10-2004
Rating: N/A
Votes: 0
|
| | Friday, March 16, 2007 - 11:25 pm: |
|
Hi Hilarius
Looks like the chart shows it was less than 2 times but more than 1.5 times.
Are you questioning the scan result or the data for IC or the 50 ema ?
ema shows 307,000 and volume on friday 9th shows 537,000
|
hilarius
Member
Username: hilarius
Post Number: 2129
Registered: 04-2004
Rating: N/A
Votes: 0
|
| | Friday, March 16, 2007 - 11:33 pm: |
|
msparks
I am unclear how the 50day ema of volume is obtained if it is above 300,000 as you suggest
Figures I've collected in Excel give a 50 day ema under 300,000 using 3 sources (including IC Charts)
Where does 307,000 come from?
That seems a high figure giving (as you say) less than 2 times average volume
With Best Wishes
Hilarius
I come in peace to share my thoughts and to shine my candle light on possible long term opportunities
|
msparks
Member
Username: msparks
Post Number: 820
Registered: 10-2004
Rating: N/A
Votes: 0
|
| | Saturday, March 17, 2007 - 05:44 am: |
|
Perhaps you calculated the Simple Moving Average
Shows 288,000 on chart with sma of volume.
|
hilarius
Member
Username: hilarius
Post Number: 2130
Registered: 04-2004
Rating: N/A
Votes: 0
|
| | Saturday, March 17, 2007 - 07:57 am: |
|
msparks
If you view the Excel worksheet above you will see that I calculated the Exponential Moving Average using data from 3 sources :-
(1) Online Broker
(2) Yahoo
(3) Incredible Charts (as per BOQ Daily Chart volumes)
All three sources gave a 50 Day EMA above 2 on Friday 9th
Thus my question why (to the best of my knowledge) BOQ appeared only in the 1.5 times scan result
With Best Wishes
Hilarius
I come in peace to share my thoughts and to shine my candle light on possible long term opportunities
|
colin_twiggs
Member
Username: colin_twiggs
Post Number: 2840
Registered: 09-2002Rating: N/A
Votes: 0
|
| | Saturday, March 17, 2007 - 01:53 pm: |
|
Roger,
For an accurate EMA you have to calculate from the beginning of the data -- not just the last 50 days. 50-Day VMA was +/- 286,000 on March 9 according to the chart; so if we divide volume of 537.8 by 286 we arrive at a ratio of 1.88.
Regards,
Colin
|
hilarius
Member
Username: hilarius
Post Number: 2131
Registered: 04-2004
Rating: N/A
Votes: 0
|
| | Saturday, March 17, 2007 - 03:08 pm: |
|
Bingo !!!
That is exactly the explanation I was looking for !!!
Thanks, Colin
Intuitively a 50 day moving average that is longer than 50 days seems an oddity .. but I guess that is why they call it exponential
I understand now that 50 days applies to part of the formula and that the length of the whole calculation is >50
So for BHP does the exponential ema go back 100 years?
Just wondering !!!
Your Ever Curious Friar
Hilarius
I come in peace to share my thoughts and to shine my candle light on possible long term opportunities
|
colin_twiggs
Member
Username: colin_twiggs
Post Number: 2842
Registered: 09-2002Rating: N/A
Votes: 0
|
| | Saturday, March 17, 2007 - 03:18 pm: |
|
Roger,
After a reasonable period of time, say two or three years, the differences become insignificant. We calculate from the start of our data which is 15 years.
Regards,
Colin
|
hilarius
Member
Username: hilarius
Post Number: 2133
Registered: 04-2004
Rating: N/A
Votes: 0
|
| | Saturday, March 17, 2007 - 06:20 pm: |
|
Hi Colin
Thanks for the various items of information you have provided ... it has been an interesting learning curve
I probably do need to make a request for a more responsive 2 times volume scan given the fact that long term price history pushed the BOQ volume spike down below 2
A spike above 2 resulted from a Simple 50 day moving average, and from my truncated "Exponential" view using a mere 50 days data
Perhaps the option of a Simple Moving Average 50 day 2 times spike would be worth considering?
It certainly would have given a most attractive trigger to buy BOQ following Friday 9th March
I was sorry to miss the chance to enter such a superb Queensland company using the IC 2 times scan which has been adopted with considerable affection by the user group looking at volume/price scans !!!
Our affection for it would have been still greater if the scan had got us into one of the best potential trades in BOQ in recent times, in an otherwise difficult market
May I organise a week, or a month ... or if need be ... a year of prayer at the Friary for a 2 times Simple Average Volume scanning option?
With Best Wishes
Hilarius
I come in peace to share my thoughts and to shine my candle light on possible long term opportunities
|
msparks
Member
Username: msparks
Post Number: 822
Registered: 10-2004
Rating: N/A
Votes: 0
|
| | Sunday, March 18, 2007 - 01:11 pm: |
|
EMA OF VOLUME
SMA OF VOLUME
|
hilarius
Member
Username: hilarius
Post Number: 2134
Registered: 04-2004
Rating: N/A
Votes: 0
|
| | Sunday, March 18, 2007 - 01:47 pm: |
|
Hello msparks
I am not sure how the SMA in your above chart is calculated
If you take the last 50 days of data in my above spreadsheet the following answers are obtained
Online Broker 50 Days Volume = 12,583,638 = 271,673 per day
Yahoo Finance 50 Days Volume = 12,684,200 = 253,684 per day
IC Charts vol 50 Days Volume = 12,857,470 = 257,149 per day
This is my evidence for 2 times average volume on Friday 9th March
I realise that IC currently uses EMA ... but a simple average scan should have >2 times average volume on my calculations
Then again I trained by counting one potato too many and having to count them all again
With Best Wishes
Hilarius
I come in peace to share my thoughts and to shine my candle light on possible long term opportunities
|
msparks
Member
Username: msparks
Post Number: 823
Registered: 10-2004
Rating: N/A
Votes: 0
|
| | Sunday, March 18, 2007 - 07:14 pm: |
|
Hilarius
I think a sma is sum last 50, and divide by 50.
Anyway according to the charts neither were in the > 2 times volume zone so the scanner is working fine and the scanner uses the ema of volume.
The issue you need clarification from Colin about is "why do the charts show different results to a manual spreadsheet calculation ?
|
hilarius
Member
Username: hilarius
Post Number: 2136
Registered: 04-2004
Rating: N/A
Votes: 0
|
| | Sunday, March 18, 2007 - 08:14 pm: |
|
msparks !!
We appear to agree on how SMA is usually calculated, but I don't understand how the SMA is calculated in your chart
We also agree that the IC scanner currently uses EMA
My point is that a genuine opportunity was missed in the case of BOQ because the 2 times EMA result seems to be too heavily influenced by old out of date volumes
I am looking for a scan that defines "2 times" by looking at relatively recent events, not relatively far distant events ... so that outstanding opportunities such as the recent BOQ price appreciation are not ignored by a more current form of 2 times scan
Since we agree on all the facts as they are (other than the mysterious SMA in your chart) can we also agree that in life there is always the chance to improve, rather than accept that what is must always be?
I hope we can
With Best Wishes
Hilarius
I come in peace to share my thoughts and to shine my candle light on possible long term opportunities
|
colin_twiggs
Member
Username: colin_twiggs
Post Number: 2843
Registered: 09-2002Rating: N/A
Votes: 0
|
| | Monday, March 19, 2007 - 04:16 pm: |
|
Roger,
I would have thought that you need a stable benchmark (e.g. 50-day EMA of volume) but greater flexibility in terms of setting the volume spike level (e.g. user input of spike level at levels like 1.8 times or 2.5 times).
Regards,
Colin
|
hilarius
Member
Username: hilarius
Post Number: 2137
Registered: 04-2004
Rating: N/A
Votes: 0
|
| | Monday, March 19, 2007 - 04:38 pm: |
|
Colin
The operating principle you've given is correct, but the limitation to EMA in practice leaves a little to be desired
The concern is that the "stable" benchmark in the case of an EMA seems to incorporate a reference period and volumes well before the latest 50 days
This appears to have the effect of increasing average volume in a way that a simple moving average avoids, at least in the BOQ example and perhaps with others also
A simple moving average in combination with the 2 times option would have given an attractive signal on 9th March for entry into BOQ
The EMA by contrast gave no such signal in a 2 times scan
I've tried persuading the user group that a 1.5 times scan would have worked but some members are adamant that 2 times is the minimum we should use, and it is also clear that BOQ had a spike of 2 times in simple average terms.
This being the case only a Simple Moving Average with a 2 times option would have provided the scan result which BOQ appeared to deserve
My concern in summary is that old volumes are having too much influence on the only available 2 times option, for our purposes
This practical problem would be overcome if a 50 Day Simple Moving Volume Average and 2 times spike scan could be offered
With Best Wishes
Hilarius
I come in peace to share my thoughts and to shine my candle light on possible long term opportunities
|
colin_twiggs
Member
Username: colin_twiggs
Post Number: 2844
Registered: 09-2002Rating: N/A
Votes: 0
|
| | Monday, March 19, 2007 - 05:08 pm: |
|
Roger,
The problem with simple moving averages is their tendency to "bark twice" -- once when unusual data is added and again when it is dropped at the end of the time period.
Regards,
Colin
|
hilarius
Member
Username: hilarius
Post Number: 2138
Registered: 04-2004
Rating: N/A
Votes: 0
|
| | Monday, March 19, 2007 - 05:34 pm: |
|
Colin
I cannot disagree ... what you say is statistically sound, but still seems to give a disappointing result
It seems that the smoothing applied in the case of BOQ led to a higher average than the simple average data, which was relatively uncontaminated as I see it, this time around
So in curing one problem another was created, namely the missing of the BOQ opportunity in the 2 times scan, even though the relatively uncontaminated volumes for BOQ seemed to call for attention when the volume doubled on a 50 day view
The question "What is double volume" is certainly a tricky one, but adding too many past volumes seems to create as much difficulty as it solves
I recognise the argument that a large spike during the 50 days has its own contaminating effect, but so does the consideration of older higher volumes during a 50 day quiet period
I can't avoid the thought that BOQ really did double its volume, but I don't have an immediate solution that would fit all cases
One of the best price jumps by a bank in recent times was missed by the user group ... and yet you offer a cogent well argued reason for rejecting the use of a simple moving average
This dilemma needs to rest at the bottom of the swamp until a suitable light bulb is found to light the way forward
You have extinguished my candle of hope, and I retire to the solitude of Evening Prayer
With Best Wishes
Hilarius
I come in peace to share my thoughts and to shine my candle light on possible long term opportunities
|
colin_twiggs
Member
Username: colin_twiggs
Post Number: 2845
Registered: 09-2002Rating: N/A
Votes: 0
|
| | Monday, March 19, 2007 - 05:44 pm: |
|
Roger,
Would a 10-day or 20-day EMA of volume be more suited to your purposes?
Regards,
Colin
|
hilarius
Member
Username: hilarius
Post Number: 2139
Registered: 04-2004
Rating: N/A
Votes: 0
|
| | Monday, March 19, 2007 - 06:19 pm: |
|
Colin
My candle just flickered back to life !!
I believe those options would be most helpful, and may well have captured BOQ in the double volume scan !!!
I would use both ... but either would be good to have
Returning to bended knees in thanks, if either or both become possible
Hilarius
I come in peace to share my thoughts and to shine my candle light on possible long term opportunities
|
ohkoolnutz
Member
Username: ohkoolnutz
Post Number: 545
Registered: 10-2005
Rating: N/A
Votes: 0
|
| | Tuesday, March 20, 2007 - 08:06 am: |
|
I see people argue that the scan was not suited for finding this single case so the scan must be at fault while ignoring the fact that the scan continues to churn out results (that could be quite profitable).
The IC scans are static and a strict framework and if a trader operates in a different framework then the scans become useless to this person. If you change the framework then you change the target group this framework provides use to.
I think this thread highlights again that different people see things differently and IC may need to move to a programming API that allows the users to write their own scans.
---
ohk
Lies, Damn Lies and Technical Analysis
|
prislys
Member
Username: prislys
Post Number: 1
Registered: 04-2004Rating: N/A
Votes: 0
|
| | Thursday, July 26, 2007 - 11:12 am: |
|
Is it 4 months since this thread hibernated ?
Hi Colin,
I'm interested in a simple indicator line to reflect (say) (i) portfolio total historical value versus an existing IC individual stock,
(ii)user forecasted price versus am existing IC generated trend line, etc...
Is this sort of user maintained data easily added to IC arsenal ?
Cheers,
prislys
|
BOQ
boq
