May

11

 Do High beta stocks tend to be leaders in various time frames of movements and do low beta stocks tend to be laggards?

Such a question carries the in-built flaw that the beta itself is not stable across time frames and hence may require a modification in studying the original question of studying it within say +/- 2 standard deviation ranges of any particular beta level.

Why such a query arose in my mind today is that the ratios of Price levels on various emerging market index futures (Korea, China, India etc.) vs the SP1 tend to reverse ahead of the SP1 at bottoms and tops on visual inspection across last 7 years or so that I inspected. Given the amplitudes or swings of movements across spans of movement ranging from days to weeks to many years is far higher in these markets (including the drift) compared to the SP1 for the last decade one has tended to make a comparison of such with SP1 as being higher beta with the beta of SP1 being 1.

An exercise in quantification of such a frame of thought across either these indices or constituents of any particular index itself vis-à-vis the index may have some utility.

One wonders aloud here, that if the situation is similar as the race between a tortoise and a hare where the hares run ahead, get pompous and lie down for rest while the tortoise does take the hares over (during beta changes) and another variant of the story not told where the hares run so quickly ahead that they genuinely are able to move only very slowly as they get closer to the finishing line while the tortoises continue to turn up at their own pace and the race to the other end of the course begins again similarly.

If one wanted to also think in terms of a third set of players, the snails running the race and never reaching any end of the race track wherein after the tortoise have reached the finishing line the hare and the tortoise start returning and the snail changes course finding them somewhere still in a new race. I would say the likes of C , AXP , BAC , JPM have been the hares, the likes of JNJ , PG , XOM have been the tortoises and the likes of MSFT have been the snails in the last big race.

I am certain that I can learn much, observing good ways of application of quantitative tools on such a theme if some of the proficient quantitative minds on the list do consider this an interesting thought to dissect it in such manner.

Phil McDonnell writes:

PhilAs usual Sushil has raised a deep and interesting question. For clarity one should start by defining two different terms. One is a leading indicator, by which is meant a stock that turns before the entire market turns. The second is the idea of a market leader. A market leader is one that out paces the market during its current move.

We recall that beta is derived from the following regression model:

(stock change) = b * (market change) + a

where b is beta and a is alpha.

From the foregoing definition of beta we can see that a high beta stock will necessarily be a market leader in the sense of moving farther and faster than the market during a given move. The alpha is really a measure of the performance independent of market risk. So in that sense alpha is really what the shrewd investor should seek.

To answer the question as to whether certain stocks turn before the bottom or turn before the top is somewhat different. For that we can look at lead lag relationships. For me the simplest way is to look at correlations at various lags. The chair recently mentioned the weakness in the Nasdaq index relative to the S&P and Dow. If we look at the movements of the QQQQ ETF 4 days ago we find that it is 23% correlated to the SPY ETF changes today. Thus the Q's act as a leading indicator. To be significant the correlation should be about 25% so its a little too weak to be considered academic proof.

Dr. McDonnell is the author of Optimal Portfolio Modeling, Wiley, 2008


Comments

Name

Email

Website

Speak your mind

Archives

Resources & Links

Search