To print article, click Print on your browser's File menu.

Go back

Posted 10/26/2000


Kenner & Niederhoffer
What's next for
Wall Street
sponsored by:
Click Here!

Extra!
The long and short of year-end momentum buying
Research shows that great performance in November and December is predicated on great performance in September and October. So here's why PeopleSoft and Paychex may pack your piggy bank for the holidays.
By Victor Niederhoffer and Laurel Kenner

Consider two Nasdaq 100 companies to whom Dame Fortune has granted dramatically different fates: PMC-Sierra (PMCS, news, msgs) and Legato Systems (LGTO, news, msgs).

PMC-Sierra was up 148% in 2000 through Tuesday, but down 15% for the past two months. Its earnings are robust, and almost every day yet another brokerage company puts out a favorable recommendation on it.

Legato, on the other hand, was down 82% on the year, but advanced 3% in the last two months. One day it's downgraded to hold by a brokerage company. Another day it appoints a new chief executive officer. A third day, it reports sales well below forecasts. And on a very bad day, the company says the SEC is investigating a restatement of financial results.

Question is, which of these two companies is a better buy right now: The one with great year-to-date performance and bad recent performance? Or the one with the terrible year-to-date performance, but relatively decent performance in the last two months? Keep in mind that the former traverses a path strewn with roses, while the other would try the patience of a saint.

Kenner & Niederhoffer
Victor Niederhoffer has traded stocks, currencies and futures worldwide for the past 40 years; he is the author of "The Education of a Speculator." Laurel Kenner is a trader and former Bloomberg markets editor. In this series of columns for MoneyCentral, they'll assess the past week's Wall Street performance and next week's prospects. Let us know what you think in the Start Investing Community.

To answer this question, we could analyze the fundamentals -- try to ascertain, for example, where earnings and sales are likely to come in relative to anticipations and current price. Or we could try to operate directly on the price and see if there are any systematic augurs in the year-to-date and recent performance.

Get a clue from Value Line
We decided to take a page from Value Line, which is in the unique position of providing both a fundamental and a technical investment recommendation for most widely held stocks. Their timeliness rankings, based on fundamental analysis, are well known; their top "Group 1" picks have consistently yielded more than 15 percentage points above the market each year for the past 45 years.

Not so well known is the performance of Value Line's technical recommendations. These are based on past price alone, and show a record that is superior even to the timeliness rankings. The methodology is based partly on separating the performance of a stock into long-term and short-term price movements.

We decided to test whether the long or the short term is more predictive, and in the process to answer the original query: whether it's better to buy a company with the performance of a PMC-Sierra or a Legato to hold for the last two months of a year.

The answer, in a nutshell: Don't buy either of them. Municipal bond

The long and the short of it
Instead, buy stocks that did well in both periods: PeopleSoft (PSFT, news, msgs) and Paychex (PAYX, news, msgs), to name a couple. (A complete list appears below.)

We arrived at that conclusion by calculating returns in 1997, 1998 and 1999 for a long-term period, consisting of the 10 months through the end of October, and a short-term period consisting of September through October.
Is it better to buy a company with the performance of a PMC-Sierra or a Legato to hold for the last two months of a year? The answer, in a nutshell: Don't buy either of them.
 We used what's known in statistics as a prospective sample: only companies that were actually in the Nasdaq 100 during the year considered were included.

For ease of calculation, we classified the performance of each company in each period by assigning it to one of four categories: top-performing 25% (1s), second-best 25% (2s), third-best 25% (3s) and worst 25% (4s).

For example, here's how we classified Broadvision (BVSN, news, msgs), using closing prices:
  • Up 590% from Dec. 31, 1998, through Oct. 31, 1999: Group 1.
  • Up 122% from Aug. 31, 1999, through Oct. 31, 1999: Group 1.
  • Up 131% from Oct. 31, 1999, through Dec. 31, 1999: Group 1.
The results were consistent in each year considered. And with 25 observations in each quartile -- and 75 in each quartile over the three years -- the chance of random results is as unlikely as the chance that spare parts in an auto junkyard might assemble themselves into a Mercedes-Benz. The following table shows what we found:

1997-1999 year-end momentum
Returns of Nasdaq 100 companies in last two months of the year, 1997-1999.
Jan.-Oct. ranking Nov.-Dec. % rise*
Best 25% 38
2nd Best 25% 25
3rd Best 25% 20
Worst 25% 10
*Note: Average Nov-Dec performance for all
stocks in sample during these years was 23%.

Sept.-Oct. ranking Nov.-Dec. % rise*
Best 25% 33
2nd Best 25% 24
3rd Best 25% 20
Worst 25% 16
*Note: Average Nov-Dec performance for
all stocks in sample during these years was 23%.

Momentum is king
In general, both the long-term and short-term results are predictive of the performance of companies during the last two months of the year. Momentum is king during this period.

More specifically, the longer period is a better predictor than the shorter period -- but both have value. In fact, they add to each other: good performance in both the short and long term was better than good performance in just one period. And bad performance in both periods was worse than bad performance in just one. (In statistics, this is described as the additive effect with no interactions, and involves adding the incremental impact of both of the key causal factors.)

How to explain the results? Maybe momentum investors kept buying the best performers, while individuals sold off losers for tax reasons.

Or maybe the price from the first 10 months of the year is somehow predictive of the actual fundamentals of the company, and these fundamentals somehow get translated into further superior price performance in the last two months of the year.

Good 8, bad 7
And now the question is, will the contestants go on and risk it all to go for a million for the year? With great trepidation we list the eight companies that were superior in both time periods this year. We predict they will show a highly superior performance in the last two months of 2000. Here's the envelope from the senior partners of the Big Five accounting firm that prepared the answers: There were also unfortunately seven losers, which are not candidates for buying: We're partial to Nasdaq 100 stocks because we believe they will show at least a 10 million percent return per century, so we do not recommend shorting the Sorrowful Seven listed above. However, we record them as candidates for eschewing vis-à-vis new purchases during the last two months of the year.



Recent Articles
• When best and worst diverge, follow the worst, 10/19/00

• Big stocks under $5: real bargains?, 10/12/00

• Revenge of the old-hearted investors , 10/5/00

more...
 The final question that emerges is how these results, which show that momentum should be bought, jibe with many seemingly similar studies that have come to different conclusions. Some of our own studies, for example, have shown that beaten-down favorites are good candidates for purchase, and indeed, we recommended such a list in our column of Oct. 19.

God is in the details
The key is in the details. The companies recommended here are Nasdaq 100 companies for a two-month holding period, from Oct. 31 to year-end. The stocks recommended in last week's column were S&P 100 companies recommended for a three- to five-year holding period. Both results are consistent. However, they have much variability and armchair quarterbacking in their conclusions.

During recent years, short- and long-term momentum of the kind that we have described has been very good. The last two months of the year in each of the last three years have been relatively good for the Nasdaq 100, and this may have contributed to the results. It's quite possible that if the Nasdaq falls in the last two months of the year, the results will be completely different.

However, we'll stick with our recommendations until further notice. The rich tend to get richer and the poor get poorer, at least during the last two months of the year.

At the time of publication, Victor Niederhoffer held shares of Novell. Laurel Kenner did not own or control shares in any of the equities mentioned in this column.




MSN Money's editorial goal is to provide a forum for personal finance and investment ideas. Our articles, columns, message board posts and other features should not be construed as investment advice, nor does their appearance imply an endorsement by Microsoft of any specific security or trading strategy. An investor's best course of action must be based on individual circumstances.