37 lines
2.9 KiB
Plaintext
37 lines
2.9 KiB
Plaintext
472 Part IV: Additional Considerations
|
||
spreads, some naked writes and ratio calendar spreads, fewer straddles and ratio
|
||
writes, and a few covered call writes. This theory would be somewhat difficult to apply
|
||
in practice, because of the massive numbers of calculations involved and also because
|
||
of the accuracy of closing price data. It was mentioned previously that a computer will
|
||
assume that "bad" closing prices are actually attainable. By a "bad" closing price, it is
|
||
meant that the option did not trade simultaneously with the stock later in the day, and
|
||
that the actual market for the option is somewhat different in price than is reflected
|
||
by the closing price for the option. A daily contract volume "screen" will help allevi
|
||
ate this problem. For example, one may want to discard any option from his calcula
|
||
tions if that option did not trade a predetermined, minimum number of contracts
|
||
during the previous day. Data that give closing bids and offers for each option are
|
||
more expensive but also more reliable, and would alleviate the problem of "bad"
|
||
closing prices. In addition to a volume screen, another way of reducing the calcula
|
||
tions required is to limit oneself to strategies in which one has interest, or which one
|
||
is reasonably certain will fit in well with his investment objectives. Regardless of the
|
||
limitations that one places upon the quantity of computations, some computer power
|
||
is necessary to compute expected return. A sophisticated programmable calculator
|
||
may be able to provide a real-time calculation, but could never be used to evaluate
|
||
the entire option universe and come up with a ranking of the preferable situations
|
||
each day. On-line computer systems are also available that can provide these types of
|
||
calculations using up-to-the-minute prices. While real-time prices may occasionally
|
||
be useful, it is not an absolute necessity to have them.
|
||
One other by-product of the expected return calculation is that it could be used
|
||
as another model for predicting the theoretical value of an option. All one would have
|
||
to do is compute the probabilities of the stock being at each successive price above
|
||
the striking price of the option by expiration, and sum them up. The result would be
|
||
the theoretical option value. These data are published by some services and general
|
||
ly give a different theoretical value than would the Black-Scholes model. The reason
|
||
for the difference most readily lies in the inclusion of the risk-free interest rate in the
|
||
Black-Scholes model and its omission in the expected return model.
|
||
APPLYING THE CALCULATIONS TO STRATEGY DECISIONS
|
||
CALL WRITING
|
||
One method of ranking covered call writes that was described in Chapter 2 was to
|
||
rank all the writes that provided at least a minimal acceptable level of return by their
|
||
probability of not losing money. If one were interested in safety, he might decide to
|
||
use this approach. Suppose he decided that he would consider any write that provid- |