Suppose that I began sometime - say, twenty years ago - with $100K invested in the S&P 500 and a few bucks
in some (almost) risk-free investment ... say a Money Market fund (MM) which pays a constant 3%.
My goal is to achieve (if I can) an annual return of 8% or 9% or 10% ... by taking money out of the
S&P Portfolio and sticking it into MM when the market increases dramatically, and putting
the MM dollars back into my Portfolio when the S&P decreases dramatically.
Uh ... well, I made a bunch of money in 1999 (76% annual gain) then lost those gains in 2000
and I promised that this wouldn't happen again.
>Promised? Promised who?
My wife. Anyway, let me continue.
Here's what I suggest:
>What's this acceptable deviation? I mean ...
- Start with $100K Portfolio, invested in the S&P 500 and
$15K in Money Market (paying 3% interest).
- We want a 10% annual gain, plus or minus some acceptable deviation.
This gives me a Minimum and
Maximum graph, as shown:
- Whenever our Portfolio exceeds the
Maximum, we sell enough stock to bring our
Portfolio back to the 10% graph (shown in
- Whenever our Portfolio falls below the
Minimum, we buy enough stock to bring our
Portfolio back up to the 10% graph.
In this example, we draw three graphs:
>Where does all this money come from?
- The desired portfolio grows at 10%.
- The Minimum portfolio is 10% less than desired.
- The Maximum portfolio is 25% more than desired.
From our Money Market fund ... hopefully.
Yes, so long as there's enough money in that MM fund.
Here's a chart of the withdrawals from our Portfolio
>I assume that the annual gain percentage of 11.2% is ...
That's the gain in my total investment:
Portfolio plus Money Market.
>And is there enough money? I mean was there?
Yes, as a matter of fact. But since I'm talking about a strategy devised today using
historical data (that ended last month!) it's fiction. However, if one wants to test a strategy
one should see what would have happened had it been applied to historical data. If it works,
then we have a good feeling about the strategy.
>But can't Monte Carlo predict the future ... sort of?
Sort of? Yes, sort of ... but then so can my grandson.
>Is he smart?
Yes, for a two-year old.
>How does this strategy do when the market gets hammered?
I assume you can maintain the 10% return.
Uh ... not exactly, but the idea is to minimize market downturns while giving up some gains.
As you can see, our Portfolio actually increased, but the
Money Market funds decreased so the gain in my Total investment over this year
was -4.7% ... but the S&P 500 was twice as bad:
I might also point out that the Volatility of our investments is smaller than that of the
market. Indeed, the annualized Standard Deviation of our investments was 11.3% (over twenty years)
compared to the S&P market which had a Volatility of 15.1%.
Yes, we calculate the Standard Deviation of the monthly gains (over twenty years), then multiply
by the square root of 12 to get an annual Volatility.
(See SD Stuff.)
>In Fig. 2 you end up with a lot of money in Money Market.
That's in case the market heads South, like, for example, if the market decreases
significantly over the next two years, here's what'd happen to our Money Market funds:
>In Fig. 4 you've got an S&P500 graph. Is that ...?
That's what would have happened with a buy-and-hold strategy.
>But the buy-and-hold did better than your strategy!
Uh ... yeah, but ...
>You started by saying you lost the 76% gains you made in 1999.
I assume that was in the S&P?
Well, no. Mostly Nasdaq and that's where this strategy does best. Here's what would have happened had I adopted this strategy when
I retired in June, 1993, where the Portfolio grew within the
desired Minimum and Maximum
and the Total Investment $$$ includes the Money Market funds
... Note the decrease of Volatility/Standard Deviation (SD)!
>How about those last two of years when the Nasdaq was out of control?
Here's the picture from April 1/99 to April 1/01:
I would have lost a few bucks, but the drastic
reduction in annualized Volatility
from 43% (for the NAZ) to 23% (for my total investments) - over this 2-year period - would
have given me a good night's sleep.
>A good sleep?
Well, a better sleep.
>I'd guess that you'd have been happy with that 14.1% annualized return, since June, 1993 and ...
>... and I suspect that your strategy is best for very volatile markets.
By the way, were you just watching your portfolio or were you withdrawing funds
in which case ...?
Yes, that's true. We should consider the effect of withdrawings funds ...
>And asset allocation and foreign and domestic stocks and bonds
and rebalancing and transaction fees and foreign exchange losses and ...
Yes, yes ...
for Part II