AAPL Accumulator, 10/29/05
Goal: Build a system that regularly accumulates shares of a stock, in this case AAPL*, within a given budget and based on the stock's historical price patterns. The goal of the program is to accumulate shares at a better price each month than simply buying by calendar date.
The first question, the baseline question, is: If I buy on the last day of the month, what will my accumulation look like? We'll compare all the other accumulations to this one.
Starting in 2003, buying 100 shares each month thru Sept of 2005, the value of the accumulation on 10/28/05:
| Number of Purchases |
Portfolio Profit (as of 10/28/05) |
|---|---|
| 32 | $103,946 |
The second question is: How many x% declines are there per year?
| Decline | Count Per Year |
|---|---|
| 6% | 5 |
| 5% | 7 |
| 4% | 7.5 |
| 3% | 9 |
| 2% | 10 |
| 1% | 11 |
The third question is: if I buy x% declines, can I beat simply buying at the end of each month?
| Decline | Number of Purchases | Portfolio Profit (as of 10/28/05) |
Beats Last Day of Month |
|---|---|---|---|
| 6% | 17 | $56,832 | No |
| 5% | 20 | 70,110 | No |
| 4% | 22 | 79,186 | No |
| 3% | 29 | 101,916 | No |
| 2% | 30 | 105,872 | Yes |
| 1% | 33 | 110,279 | Yes |
The first optimization is: what if I buy x% declines and if I there hasn't been an x% decline by the last day of the month, I buy on the last day of the month?
| Decline | Number of Purchases | Portfolio Profit (as of 10/28/05) |
Beats Last Day of Month |
|---|---|---|---|
| 6% or LDOM | 33 | $105,406 | Yes |
| 5% or LDOM | 33 | 105,327 | Yes |
| 4% or LDOM | 33 | 105,114 | Yes |
| 3% or LDOM | 34 | 109,486 | Yes |
| 2% or LDOM | 34 | 110,615 | Yes |
| 1% or LDOM | 34 | 111,881 | Yes |
The first conclusion is: This is how you invest in stocks - with a budget and regular allocations as opposed to one shot lump sum win or lose plays.
The second conclusion is: It is better to regularly catch a small decline than to irregularly catch a big decline.
The fourth question is: What if my budget is $2000/mo instead of 100 shares/mo and I buy on the last day of every month? (Note that the average purchase price in the above scenarios is about $20/share or $2000/mo)
| Number of Purchases |
Portfolio Profit (as of 10/28/05) |
| 32 | $179,000 |
And buying after an x% decline or on the last day of the month:
| Decline | Number of Purchases | Portfolio Profit (as of 10/28/05) |
Beats Last Day of Month |
| 6% or LDOM | 33 | $185,503 | Yes |
| 5% or LDOM | 33 | 185,903 | Yes |
| 4% or LDOM | 33 | 185,117 | Yes |
| 3% or LDOM | 34 | 191,018 | Yes |
| 2% or LDOM | 34 | 192,323 | Yes |
| 1% or LDOM | 34 | 192,828 | Yes |
The third conclusion is: It is better yet to regularly catch a small decline w/ a fixed dollar allocation than a fixed share allocation.
The fourth conclusion is: The actual AAPL Accumulator algorithm -
if (this is the start of a new month) then----
set buy flag = true
if (buy flag == true) and
((close today is less than 1%
below the close five days ago) or
(this is the last trading day of the month)) then
buy (monthly allocation budget)/(share price) shares
of AAPL on the close
set buy flag = false
*What got me thinking about AAPL was thinking about the strategic implications of AAPL's move to Intel processors:
If the biggest barriers to buying a Mac - real or imagined - are "there's no software for the Mac" or "I'm dependent on program x and program x isn't available for the mac", doesn't the move remove those barriers?
And if it does, in a PC world very tired of spyware, viruses, msft taxes and general inelegance isn't there the some potential for a _huge_ coup: the world might just switch.
Or, if not the world, then 1/2 or 1/4 or 1/8 of the world. Which, at a minimum, gives Apple a 20% market share, 3x its current run rate.
And, strategically, that would be worth a try, especially if the try could be made w/o too much risk and especially to a company that has gotten a taste of being a pseudo-monopoly.
From strategic and risk/reward standpoints, the move seems quite brilliant.