Here is what I'm asking of the software Guru's. If I had three PC's in front of me and I set up a stock on all three with the same data. Would I see the same Buy & Sell recommendations across the board?
Hi Keith,
I'm confident that all 3 products will produce the exact same trades if the exact same prices are plugged in. We created pca using the basic 'by the book' algorithm and it remains intact to this day.
On pages 64-71 of the book, Mr. L calculates a hypothetical price progression of 10-8-5-4-5-8-10, repeated 15 times. We call this the 'test' pattern and include it in every copy of pca.
If you turn to page 71 and look at the results, after 15 cycles of this pattern (61 trades), you will see that the values match up to the 'test' in pca.
stocksystem.com
Since there are no values on the last line in the book, we compare them to the second to last line corresponding to the last price of 8.
Shares Owned
book-81,348
pca-81,351
Current Trade Advice
book-sell $260,601
pca-sell $260,607
The minor differences are are attributed to rounding as mentioned previously. The demonstration in the book assumes that a dividend is added at every price point, however in pca, the user must add dividends to the beginning cash as they come in. That is why the cash and portfolio value do not match exactly.
Originally, when we were creating and testing pca, we found that the cash reserve would easily get depleted in many instances and decided to create a filter that would quit issuing buys if there was not sufficient cash reserve. This feature is nice because an investor can choose a fixed amount of capital to commit and never be asked to put in more. In the case of something like a Boston Chicken, this could be quite beneficial. Simply adding cash on the Control Panel of PCA will allow the user to see the buy advice even if the cash reserve is depleted.
PCA assumes a 50/50 split between cash and equity by default. Again, this is easily adjustable, but extensive testing showed a large number of stocks that depleted a 50% initial cash reserve. I even wrote a summary of a test that shows that with a 50% beginning cash reserve, a 70% stock decline would deplete that cash.
stocksystem.net
Lately, with many stocks well off their highs, I'm more inclined to go with the IW or even a 33% beginning reserve, however, I always like to err on the side of being conservative, especially when it comes to new AIMers. A quick glance through the charts shows many cases where the cash reserve gets fully depleted even at 50%.
It would be interesting for someone to test of all 3 AIM software products, entering identical price data, and report the results. I'm certain that the original algorithm is intact in the calculations of each. It's the implementation that differs.
The dialogue here lately is great and I see now that it's important that I visit here more often. I agree with the comments that these issues are very important as AIM gains momentum in the investing community. Everyday, I must start from scratch explaining AIM to new users. I certainly wish the 'book' was distributable online! Then again, most people don't even want to read the help in the software....
Anyway, thanks for bringing up such issues. We have a pca here that is guaranteed to raise the bar to a whole new level. Hopefully we will get the finishing touches on it soon. The idea of distributing over 5,000 copies of a free upgrade that isn't bulletproof scares the heck out of me! How'd ya like to be on the receiving end of those emails? We will do the best we can to get it released, but won't rush perfection.
I should stop now and read the rest of the posts here. I never thought I would need a calculator to follow along!
Best Wishes,
D1 |
