Most market/sector indexes are weighted by capitalization, which means that the largest cap stocks exert the most influence on the index. For example, I once examined the S&P 500 Index and found that the top 50 stocks (ranked by market cap) actually represented about 70% of the entire S&P 500. What this means is that the remaining 450 stocks have very limited influence on the index.

For several years Decision Point has been calculating unweighted indexes for the S&P 500 and Nasdaq 100 Indexes. Equal weighted or unweighted, the basic methodology is the same -- the index changes are calculated as the average percentage change for every stock in the index. In the case of the Nasdaq 100, add up the percentage change of all 100 stocks in the index and divide by 100.

Since each stock in the index is supposed to carry equal weight, positions must be re-balanced periodically so that they are remain equal. With Decision Point's unweighted indexes we assume daily re-balancing, but as a practical matter, the S&P 500 and Nasdaq 100 Equal Weight Indexes are re-balanced quarterly. Surprisingly, there has not been a significant difference in the results of daily or quarterly re-balancing.

Below is a chart of the new Nasdaq 100 Equal Weight Index compared with the traditional Nasdaq 100. The bottom panel is a ratio chart that shows the relative strength of new index to the old. When the ratio is rising, QQEW is performing better than the NDX, and vice versa.

But wait! If QQEW just began trading, where did all the historical data come from? We created historical data for QQEW by adjusting the data from our Nasdaq 100 Unweighted Index. Obviously, we cannot claim that is an accurate representation of historical data for QQEW, but my attitude is that it is way better than nothing, and it is probably closer than we think to what reality would have been. With this estimate of prior performance, we are able to get a technical perspective much more quickly than if we had to wait for sufficient data to drive the indicators and models.

In the end I prefer equal-weighted indexes because they perform much better than their cap-weighted alternatives. For example, the Nasdaq 100 advanced 114% from the October 2002 bear market low to the recent highs. During the same period our unweighted Nasdaq 100 Index advanced 178%.

The secret behind this superior performance has to do with the fact that the smaller-cap stocks in an index usually perform better than the large-cap stocks; however, one should always be aware of the changing relative strength between the unweighted index and its cap-weighted version. Bear markets usually cause small-cap stocks to under perform large-cap stocks, so equal weight indexes may be more dangerous than the cap-weighted alternatives.

© 2006 Carl Swenlin
Editorial Archive

CONTACT INFORMATION
Carl Swenlin
President
DecisionPoint.com
Redlands, CA USA
Website  l  Information