|
PureBytes Links
Trading Reference Links
|
Finally completed the first milestone of code ...
and got some results.
The tests were done with QP2 industry monitors
(used 196 of them) and Market indeces (21 indeces).
The strategy was based on buying the top rated
issues as measured by the percent increase over prior period.
The top issues were bought. The number of buys were
influenced by the percent of issues with positive performance. For example, if
90% of issues had positive performance, 6 issues were bought. If Less than 20%,
no buys weer made. In between, the number of buys changed
correspondingly.
Issues that were purchased and stayed in the top
15, were held, if they were not in top 15, were sold.
The test results were quite dissapointing...
essentially a small negative number.
The equity curve basically followed the broad
market and my data goes only as far back as early 2001, so it was a bear matrket
throughout and I only tested longs.
While the testing is limited, it is clear that can
not buy based on performance alone, which is essentially a measurement of
momentum.
As we have seen, momentum frequently means you aere
close to a top....
The % P/L is also not accurate because AB allocates
funds to all issues tested, while only a small fraction of the issues are
traded. So the actual P/L (in this case losses) would be magnified, perhaps
significantly.
Testing shorts would probably provide positive
results for the same time period, but it does not look like an efficient
strategy.
Combining performance with a broad market
measurement, such as New Highs / Lows or number of advancing issues should
provide much better control of timing.
Ara
Yahoo! Groups Sponsor
ADVERTISEMENT
Post AmiQuote-related messages ONLY to: amiquote@xxxxxxxxxxxxxxx
(Web page: http://groups.yahoo.com/group/amiquote/messages/)
Check group FAQ at: http://groups.yahoo.com/group/amibroker/files/groupfaq.html
Your use of Yahoo! Groups is subject to the Yahoo! Terms of Service.
|