A.I.M. Users Bulletin Board

[SFSecurity]

Hi RT,

Being a programmer, I have most of this automated.

Lucky dog.

Thanks for the tip about seekingalpha's Beta data for ETFs. That helps a lot.

The High/low range didn't seem as significant as the implied volatility.

Volatility is key, of course. In the 4th edition (which is exactly that same in the in the two previous editions as far as I can tell) the range from $10 down to $4 and back to $10 is all 20%, except for the step from $8 to $5 which is 37.5% on the down side and there is a buy order for each step. But on the up side, from $4 to $5, 125% there is no sell order but between $5 and $8, 160%, there is a sale and between $8 and $10, 125%, there is another sale but because of the increased Portfolio Control from the previous buys there is no buy from $10 down to $8. It takes the move to $5 to have another buy.

Given this, range is but a partial key to buys and sells when using BTB 50/50, especially since Lichello doesn't set a minimum percentage stock sale, only a minimum dollar amount for a trade, different from the way we do it now. If we set smaller minimum buy and sell percentage then we would get more action overall but the speed (length of time for a change) makes all the difference. You can test this by taking Lichello's range but instead such quick changes such that you get two complete cycles in 13 months you can use $10, $9, $8, $7, $6, $5, $4, etc. This way you do get the first buy at $8 and you run out of cash with a buy/sell of 10% using the BTB 50/50 cash/stock and no minimum shares. If you do this You get a lot more action. I did 3 cycles and found that it was much more action, but less profitable than Lichello's example, plus you run out of money at a $4 price in the second cycle and at the $5 price going down in the third cycle. Plus the slower change - 3+ cycles for Lichello in 25 months but only 2 complete cycles over the same 25 months with the $10, $9, $8, $7, $6, $5, $4, etc.. The rate of change in Lichello's is key to to winding up with $20,200 versus $19,375 and it appears to get worse further on. I did it by hand so I could be sure that I was copying Lichello exactly but ran out of energy after 42 months. Lichello is waaaay ahead by the the end of the table in the book. So that demonstrates rate of change, or volatility, combined with enough range, is key.

So then I went through the BareAIM.xls spreadsheet and added a column for Interest like Lichello used in the book and entered all the items listed in the book as well as the $10, $8, $5, $4... sequence to confirm the accuracy of the spreadsheet, well, not quite. The error is 0.2632554% lower than the book's table, $1,006,321 versus $1,009,281. It appears that the difference is due to accumulated rounding differences. For example, in year 1, the first trade of 75 shares is the same but the second is 428 in the spread sheet but only 427 in the book. This kind of difference accumulates so the last trade in the book is for a sale of 32,575 shares while the spreadsheet only calls for a sale of 32,307 shares, a bit over 8% difference. But overall results are quite close.

The interest in the book's table contributes quite a bit to the overall results. In the spreadsheet, without the interest the end result is only $982,693.40 versus $1,006,321 or ~2.4% extra with interest included.

Overall quite fascinating.

Best,

Allen