Register for free to join our community of investors and share your ideas. You will also get access to streaming quotes, interactive charts, trades, portfolio, live options flow and more tools.
Hi AIMster,
The issue with "average cost" is that it can be 0 or even negative.
For example, if I purchase 10 shares for $100, my average cost is $100 / 10 shares or $10 per share. If these shares then rise to $100 and I sell one share, my "average cost" then becomes $0 / 9 shares or $0 per share.
If I sell 2 shares at $100 per, my "average cost" becomes negative $100 / 8 shares or -$12.50 per share. Similarly, a case could be made to decrease "average cost" when dividends are paid out (since dividend payments decrease the value of the company, a dividend event could be viewed as selling a piece of the company and paying out the proceeds to shareholders).
From one perspective that makes sense, but to me, that perspective is attempting to combine profit and average cost into one number when they should really be separate.
Your average cost should always greater than zero. It might be a very low number (such as 2 cents per share), but it should never be 0 or less (unless someone gave you the shares or paid you to take them I suppose, but that opens another can of worms). The other number, your profit, will take into account the realized gains you make when selling.
Depending upon one's notion of accounting, the proceeds could be used to lower the "average cost" but from a tax accounting point of view selling does nothing to change the previous average cost. My guess is that Mark has chosen to use the "tax accounting" model in this regard.
Thanks, Tom,
Makes sense from that perspective. I was looking at it more from the absolute sense of if I had 100 shares, sold 10, the cost would be changed by dividing the 90 into the remainder of the sum of the cost(s) to get a per share value that way. If one were to derive such a model you would have to somehow or other tell AI from which tax lot, so to speak, the sale was made from, something that it doesn't do now. Models being FIFO, LIFO, greatest gain, greatest loss, at least as far as my ancient MYM software allows for. All depends on how you want to slice-and-dice. I suppose he could add a field for "tax basis cost" and another for "average cost," but maybe I'm building version 4.x at this point, eh? <grin>.
Best,
AIMster
Hi Tom/AIMster,
Yes, Tom is correct. In AI, selling does not change the previous average cost per share, just purchasing.
Hi AIMster, Re: Average Cost/share...................
Depending upon one's notion of accounting, the proceeds could be used to lower the "average cost" but from a tax accounting point of view selling does nothing to change the previous average cost. My guess is that Mark has chosen to use the "tax accounting" model in this regard.
Only buys will then change the average cost going forward (assuming the price/share is different from the current average).
Is this right, Mark?
Best regards, Tom
Hi, Mark,
Not sure why I didn't notice this before, but from last week's guesstimated buy, I actually got a sell recommendation. What seems to be a minor issue is that the average cost of shares didn't change after I posted the sale. Does it recalculate the average cost only on buys? Again, not a major item, but...
Thanks!
Best,
AIMster
Hi Glenn,
No problem. I actually have enough beta testers right now.
I'm working on a new website which will have one or more video demonstrations. So you should be able to see everything the new PI does.
And yes, I do plan special offers for anyone who has any of my software packages (including VSS).
Hello again AIMster,
This was actually put into AI some time ago but taken out for the actual release because the colours didn't look good. I believe I used Green for buy, Red for sell and yellow for hold (basically traffic light colours).
I'll revisit this again when doing the next service pack and see what I can come up with.
Hi AIMster,
I just tried what you describe and I see something a bit different. I can update the first portfolio to Sept. 17th. It appears to work fine. However updating subsequent portfolios is as you described. Definitely a bug I'll have to fix.
I'm surprised nobody else has reported this before, perhaps nobody updates manually and backdates at the same time.
Thanks for the report.
Mark,
I probably do not have time to do a proper beta test but I may be interested in a demo of PI at some point. Projecting forward, do you anticipate that you might offer an "upgrade" option for VSS users who may want to switch to PI?
Mark,
Yet another cool feature would be to have the recommended action field change colour as per the recommended action - green if a sell, red if a buy and some neutral color for hold. Gives a psychological reinforcement to the pending action. Just a thought..
Best,
AIMster
Hi Mark,
A small possible "bug." As you are aware, I was away on vacation earlier this week and unable to get to AI. However! I wasn't going to let the swoon of the market go unnoticed, so I "guesstimated" as the market was making new lows. Now home on Saturday, I'm going to update AI with my guesstimates. Rather than just enter the transactions as buys, I've been loading the end prices from 17 September, the day I made the purchases to force AI to give a BUY signal which it would have done had I been home!
Here's the 'bug' - when manually updating prices in multiple-security portfolios, it defaults to the current date. In each case I over-rode the default from the 20th to the 17th. Yet, when all the updated prices post, in the transaction line they're posting with the current date of 20 September. So, when I go to enter the buy transaction on the 17th, the software stops the entry due to the presumed buy being earlier than the later, though incorrect price update transaction. So I'm dating the buy to match the 20th.
What would be nice is if the multiple price update would not only post correctly, but carry forward the first date that you enter as you enter each price, rather than having to change the date each time.
Thanks for your consideration. Version 3.0, SP6
Best,
AIMster
Hi Jimjih
>>>>why would these be riskier?
dont they follow the russell 2000<<<
They are riskier in the same way mortgage derivatives are riskier than if a bank originated and held its own mortgages.
They move twice as fast ..... burning through your cash on the downside.
In the short term they follow the index but not over longer periods of time due to a number of reasons I can not explain but including higher fees , the fact that if something goes down 50% it has to go up 100% to get back to even.
Don't even think of using one fund to AIM and one as the CASH! BIG BIG mistake!
I suggest a more diversified portfolio of large value, small value, forien, REIT, Bond using ETFs
Toofuzzy
Hello Jack,
The funds TRY to follow the Russell 2000 index, but they don't actually hold the underlying index's components. They use proxies. Here's the description for TWM...
"The investment seeks daily investment results, before fees and expenses, which correspond to twice the inverse of the daily performance of the Russell 2000 index. The fund normally invests 80% of assets in financial instruments with economic characteristics that should be inverse to those of the index. It may employ leveraged investment techniques in seeking its investment objective. The fund is nondiversified."
Note that they try to achieve 2X the inverse (so that's inherently riskier right off the bat), they may use leverage (again, riskier) and the fund is nondiversified (much riskier).
UWM uses similar tactics.
uwm/twm
thanks for the reply
i did post at the other board
why would these be riskier?
dont they follow the russell 2000
Hello Jack,
Welcome to AIM.
Yes, conceptually these two ETFs should work well with AIM. However they are far riskier than solid stocks or index funds.
If you were going to AIM these two, I would suggest you have enough funds set aside for the long haul and start with a conservative model (i.e. 50% cash). You would also want to AIM these in separate portfolios so that you can capture all of their volatility (if you put them into one portfolio, you'll effectively diversify away the volatility that AIM thrives on).
I also seem to recall that someone (it might have been Tom Veale over at www.aim-users.com) did something similar a few years ago (i.e. AIMing 2X and inverse funds) and ran into a problem (I can't remember exactly what it was). However you might want to post your question on the main AIM board here --> http://investorshub.advfn.com/boards/board.aspx?board_id=949
I'm sure you'll get some additional comments on the upside/downside.
Hello
i am new to this board and to aim investing
so you think it would be possible to just trade uwm and twm
with this aim system??
thanks
jack
Google Gears looks interesting. Thanks for the link.
A better scenario would be to install AI on the thumb drive. This would require a different licensing model where the license is not tied to the hardware.
You are correct on both counts. The big issue is the licensing. Not tying the software to the hardware makes it much easier to circumvent the licensing. And the more expensive the software, the higher the potential that this will occur. But still, it's something to think about.
However copying the database file will work right now -- not the idea I had initially envisioned when giving out free trials, but one that will work in a pinch.
I had actually thought of an online version before (but the trouble with an online AI is that I would then be responsible for everyone's data -- and that's a business I definitely don't want to be in). However, what you suggest would get around that if the site where you uploaded your aidb.mdb file was in your control and only the shell was in my control.
This sounds like something Google Gears is designed for.
http://code.google.com/apis/gears/
Another alternative would be to put your aidb file on a thumbdrive (or email it to yourself) and take it with you.
A better scenario would be to install AI on the thumb drive. This would require a different licensing model where the license is not tied to the hardware.
I see that Openoffice evolved from StarOffice. Back about 10 years ago, I had downloaded and played around with StarOffice. I liked it, but it wasn't 100% compatible with Microsoft Office and I already had MS Office. So I stopped using it.
If anyone is successful in using BASE to generate user defined reports from AI, let me know.
Openoffice csn be a little querky, but usually if you google or get on their forums you can find what you need.... The best part is that it is free....
Hi Ex,
AI's mdb file was not created with Access and does not contain queries or forms. Therefore it sounds like OpenOffice will do what AIMster wants (and it's free!).
That really opens an entirely new way to use AI. I imagine you could merge AI data with other programs' too.
Thanks for the information.
Hi AIMster,
I'm writing down all of your suggestions.
I had actually thought of an online version before (but the trouble with an online AI is that I would then be responsible for everyone's data -- and that's a business I definitely don't want to be in). However, what you suggest would get around that if the site where you uploaded your aidb.mdb file was in your control and only the shell was in my control.
Another alternative would be to put your aidb file on a thumbdrive (or email it to yourself) and take it with you.
Then if you really need to access your data (e.g. on days like today), just download the trial version of AI on any computer, copy over your data and you're set. When you're leaving for home, copy the data back to the thumb drive and then back to your home hard drive.
Hi, Mark,
Another suggestion. I'm writing this from a friend's computer on Staten Island, going down to Delaware later today. Unfortunately my AI software is quietly resting on my down computer at home, waiting for my return. With the market going to the proverbial hell-in-a-handbasket today, it's altogether likely that some buy points may well get hit.
I know some use GTC orders, but I prefer the commission-free transactions available on foliofn. So, the idea bulb lit up as to the ability for you to have a web-based version of AI so that those of us on the road, either for business or vacation could still access the functionality.
I'm thinking of a site where one could upload/download one's aidb.mdf file to and there would be a shell that would let you process the file as if it were on your local PC. Might not need all the bells and whistles of the at-home version (otherwise what's the point of the standalone version). You could get the data for say a usage fee from the subscription service that you use VSS from. This way we can keep monitoring our systems and take advantage no matter where we are as long as we can access the web.
On the other hand, I get home Friday, so maybe the cash reserve will be even better deployed at the end of the week than the beginning!
Thanks!
AIMster
Aptus, Quicken won't read an MDB (Access file).
I am 100% sure that you can open and used MDB files within OpenOffice database program "BASE". If the MDB file is truly created with Access and has queries/forms etc. those won't work, but you can access the data files. I used to use some tax software that was kept in MDB files and I was able to open it and create my own reports.
Open Office is free:
http://www.openoffice.org/
It will open and save any Office file. I have ditched MS Office except for one Access application I have.
Thanks for the suggestion Tom. I'd be curious to know if Quicken can read MDB data files. If so, that would open up AI tremendously.
Please let me know if you find anything.
Hi Mark, There are some report functions inside of Quicken that might be of value. I'll take a look and see what's available.
It would require owning a copy, but that might not be a burden if one were also using it to generate cap. gain docs for tax filings, etc.
Best regards, Tom
Hi Glenn,
I still have to finish writing the explanations for rebalancing and walk forward testing as it applies to PI 3.0. As soon as I finish I'll make it available.
In the meantime, you can Google for "walk forward testing" and you'll find lots of information on it. One example is here --> http://www.amibroker.com/guide/h_walkforward.html
The rebalancing engine uses standard rebalancing methods but is a bit more complicated to explain because it operates on 2 levels (long-term and short-term), so I can't point you to anything that fully explains it right now. However I will have a full explanation in the next few weeks.
I've set up a blog here --> http://aptusblog.wordpress.com/ where I plan to post the explanations. I'll also post an update here when the blog entries are ready.
Thanks Mark. Could you describe or point to information about the rebalancing functionality and the walk-forward testing?
Hi Glenn,
Computers today are very fast and they keep getting faster.
However there are certain bottlenecks that can cause significant delays. One such bottleneck is accessing data over the Internet. Until these data are returned, the software can't really do anything. Compared to local data access, remote data access is EXTREMELY slow (the interesting thing is that "slow" is relative. 15 years ago if you told someone you could analyze 8800 stocks in a few hours, they'd have thought that was fast).
By implementing some algorithms to minimize the number of network data requests and using caching techniques, more of the processing can be done locally and that's where the vast majority of the performance improvement comes from. A small percentage comes from optimizing the code.
As it turns out there are other techniques that can be used to further increase performance. Currently the software uses a synchronous pattern (i.e. it has to wait for a response before proceeding). Changing to an asynchronous pattern and taking advantage of today's multi-core processors, using threads, would further increase performance. However processing 8800+ stocks in 1 minute 20 seconds is fast enough so that this change is not currently necessary. Once foreign support is added (e.g. Canadian and European stocks) and the number of tickers increase dramatically, then I'll most likely implement the asynchronous pattern.
PI 3.0 will be subscription-based but it is a superset of VSS. Basically it includes everything VSS includes (i.e. the fundamental analyzer and the asset allocator) plus a portfolio manager (similar to AI's), a classic rebalancing engine, News reader, historical analyzer (that supports walk-forward testing) and a new short-term rebalancing function that I've been working on for the past year.
Mark,
A couple of questions. Could you explain how you get such a speed increase over VSS? What about subscription fees? Is this basically VSS plus the fundamental analyzer found in AI?
Hi AIMster,
Yes, it will. Shortly after the PI product is released, VSS will be upgraded to use the PI engine.
Mark,
Does this new PI version imply that VSS will get the same new faster, more better, warp drive "engine" as well?
Thanks!
AIMster
Hi Everyone, I'm looking for a few beta testers for my Pragmatic Investor software upgrade. If you're interested, please send me an email (mhing@automaticinvestor.com).
I plan to have the beta version released next week. Although its listed as an upgrade, it is head and shoulders above the current PI version. It also incorporates the Value Stock Selector functionality, but is MUCH faster (currently VSS takes anywhere from 17 minutes, on a new machine, to a few hours, on single core older machines, to scan 8800+ stocks. The version built into the new PI takes about 1 minute and 20 seconds on an OLD machine).
As usual, if you're a beta tester, you'll receive a free copy of the software when its released.
If you have any questions, just let me know.
Thanks.
Hi AIMster,
I'm not familiar with any report generators, but I did a quick Google search and some low-cost/trial versions came up.
There was one (a bit expensive at $359) that looked like it could do everything -- queries, reports, charts, graphs, the works. However the low-cost ones seem to have the functionality you suggested.
Unfortunately I can't recommend any one in particular since I don't use them. However, your idea means that AI users aren't limited to the canned reports and can analyze their data in almost unlimited ways. I'm not sure why nobody thought of this before.
If you find a free or low-cost generator, let me know. I'd be interested in playing around with it. Perhaps sharing this type of thing will appeal to AI users more than sharing models.
Hi, Mark,
Is there perhaps a freeware report generator that we can use on the AI database to create individualized reports? I'd like one that would list the holdings and their tops and lows of their respective hold zones. That way I might be able to keep an intraday eye on if things are getting close to one end or the other. Rather than take more of your time, if there's such a program out there, we could write our own. Perhaps they could be exchanged between users much as you'd thought various AI configuration models would be.
Thanks,
AIMster
Hi, Mark,
Thanks for the prompt reply. No performance issues and my file's not anywhere near being excessive. It was just more of a theoretical "gee, do we really need to keep ALL this stuff" type question. The database I manage at work which runs under uniVerse (a cousin of Pick) and it's is very good on accumulating data, getting rid of it, not so much. Still, that we can back up the whole thing on to just a little over 1 DVD [in ZIP archives] is a testament to the efficiency of the Pick structure. Basically it works like a spreadsheet, though one can have multiple values and even subvalues within a single "cell." Everything's variable-length data, so the data within each record will only use the space of the length of the data. This gets around the need for predefined field lengths. Anyway, before I go on and on...
Thanks,
AIMster
Hi AIMster,
I'm a big fan of keeping all my data around.
As you say, it's probably not too often someone will go back to the start of their portfolio, but the odd time you want to can be quite valuable. Nobody would undo to the first transaction (doing this would delete all transactions that came after), but I can see people wanting to go back in time to see what they were doing and what their portfolio looked like at that point (and don't forget, the historical Notes field is accessed via the Transaction History list -- and notes are a good way to learn from your past as they show how your investment thinking evolves over time).
BTW, how large is your AI database file? The software usually does a compress on the database each time you exit so that helps keep the file size in line.
Regarding performance, there is a slight hit on VERY large databases, but I don't think it's anything an AI user would notice, even with many years of data. At the end of the day, the AI db, even used to its fullest, is not really that large relative to some other applications.
Even without getting 500 GB for $80, I don't think disk space would be an issue (I just purchased a 750GB internal drive for my iMac for $120 here in Canada -- we're a bit more expensive up here. Still, I remember paying $750 for a 5.25 inch floppy disk drive for my Radio Shack colour computer back in the early 1980s, so $120 is mind-boggling cheap).
Are you noticing any performance issues?
Regards,
Mark
http://www.AutomaticInvestor.com
http://www.ValueStockSelector.com
http://www.PragmaticInvestor.com
Hi, Mark,
Another suggestion for your Real Soon Now list. <grin> One thing with the AI database is that it keeps growing and growing and growing. Granted disk space is cheap enough these days, I recently got an external drive 500Gb for $80, but after a point, at what point would a person ever need to 'undo' to the first transaction? Or how many go back to see where things were a couple of years ago?
So, what might be worth adding at some point would be a purge function that would make the current line in the history as the starting line and add new history going forward from that point.
Such functionality would let users keep the history trimmed to whatever length is reasonable for them. Invoking a mandatory backup before the purge might protect people from an "oops" moment. Nothing to have to do right away, I don't think, but might be useful. Besides, does the Access "core" you're using have any performance issues as the number of transaction lines gets longer?
Thanks for your consideration.
Best,
AIMster
Seems to work.
Thanks.
The VSS problem should now be fixed too.
Please try it again and let me know how you make out.
Thanks for fixing this. It is good to have the Fundamental Analyzer functionality back in AI.
Any update on the VSS problem?
There is a new Service Pack (SP 7) available from --> http://www.automaticinvestor.com/ai30setup.exe
Download and install it into the SAME folder where your current version of AI is installed (ensure you back up your aidb.mdb database file first in case anything goes wrong).
This should fix the problem with the FA (which was caused by a data problem, but not a Yahoo! data problem. AI retrieved its 10 Year T-note rate data from another site and this caused the problem. I've changed it so that the T-note data is now retrieved from Yahoo! Finance's site).
I have both version 2.0, Service Pack 2 and version 3.0, Service Pack 0 installed. I get the same results with both.
Yes, that is a problem. Another user just brought that to my attention. I'm looking for the problem right now. What version and service pack of VSS do you have installed?
Regards,
Mark
http://www.AutomaticInvestor.com
http://www.ValueStockSelector.com
http://www.PragmaticInvestor.com
Apparently I was mistaken in thinking that the price/cash flow data was a problem. I've spent some time looking at this and it appears to have never been retrieved in any version of Automatic Investor. In fact I have a note saying this should be implemented once the Yahoo! data supports it (I suppose I should have removed it from the GUI rather than saying it was not available).
Therefore price/cash flow is not the cause of problems you see.
I'm running some FA tests now to see if I can spot any problems. So far the data retrieved for MSFT appears to be correct and it is rated at 1.50 (so it's a big drop from last year).
I'll try some other large and mid-cap tickers and let you know what I discover.
Thanks.
Should I be concerned that ratings and target prices have not changed for any stocks since May 31, 2008?
For anyone else who has VSS, the answer is that if you purchased it prior to July 2007, you are not on a subscription but your data is just as valid (basically you are grandfathered and receive the updated information at no charge).
For those who purchased after that date, you're on a subscription (either annual or monthly).
Never mind. I found the answer in some e-mail from a year ago.
However, I find it a bit disconcerting that the target prices and ratings have not changed from a report I had saved from May 31, 2008 to one done today (August 24, 2008). I would expect the target prices to change as the reported earnings change.
When I purchased VSS it was not a subscription based product. I was not aware that it had changed to a subscription based product. So, when I run VSS, am I getting accurate data or stale (worthless) data?
Hi AIMster,
As usual your suggestion is very good, but there are 2 issues right now.
1) VSS is currently a subscription-based package because it uses data that I have to continually update on one of my servers. However AI is not subscription-based because it retrieves its data directly from Yahoo! Finance. So to implement VSS-like functionality in AI would either mean charging a recurring fee to AI users or having VSS users subsidize AI users. Neither of these options are good ones at the moment.
2) I'm in the process of finishing a major update to the Pragmatic Investor software and thus my time is extremely limited.
However I like your idea. Of course the work-around is to use VSS to create the list of stocks, export it to a text file and then import this file into AI and analyze it. Not the seamless solution you're proposing, but it would work and it's not too difficult to do. The one thing missing would be the automatic sort, but that can be eye-balled.
Volume | |
Day Range: | |
Bid Price | |
Ask Price | |
Last Trade Time: |