Monday, December 21, 2009

'Leaders of the Pack' Share Basket

in our last post we had a look at PIE, the 'Technical Leaders' emerging markets ETF. I concluded the post by saying that the fund had already picked out the 100 most overextended emerging markets stocks for us. it made me think that we could refine that list of 100 by making a portfolio of the most overbought stocks within the overall list. say, take the ten hottest stocks in PIE and make a sub-index out of them. isolating these 'leaders of the leaders' and tracking their combined performance might give us insight into health of the broader market.

then I realized that most of the stocks in PIE didn't trade on US exchanges, so I figured it would be too much trouble to create the type of basket I had in mind. instead, I turned my attention to PIE's domestic cousin, PDP, the PowerShares DWA Technical Leaders ETF.

in Reminiscences of a Stock Operator, Larry Livingston (the fictionalized Jesse Livermore) looked for the leading stocks in the market to drop sharply and not come back as evidence that the bull market was coming to an end. (I'd quote the passage, but my copy of the book is on a different continent than I right now.) This line of observation should be just as useful for today's market as it was for the market in Livermore's day. the trouble is that there are so many different stocks these days, so how do you reduce the 'leadership' to a manageable number of symbols?

well, this PDP fund holds 100 of the most relatively strong US listed stocks. I don't know exactly what their criteria for relative strength are, but I doubt the criteria are anything to argue over. So every three months the guys at PDP put 100 stocks in the fund, and each stock is weighted equally at the time of the quarterly constitution. but as the quarter wears on the leaders within the 100 gain higher weightings.

so what I did was make a sub-basket out of the top ten holdings of the fund as of last Friday. we can infer that these ten stocks have been the top performers of the 100 components since the last time they modified the fund's holdings.


so we can recognize these ten stocks as the leaders of the broad (US) stock market over past several months. taken together they represent the essence of the bull market. they are all that is good and right with the world of investing (viewed in hindsight at least).

so now we have our list of stocks to include in the 'Leaders of the Pack' basket, but we don't yet have a formula for combining them into a collective unit. I chose to combine them on a simple share price basis. in fact the basket is simply the sum of all ten prices. the main reason I did this is because I am a total ignoramus when it comes to working spreadsheets. so in the charts that follow keep in mind that the most expensive stocks (like AAPL and PCLN) exert a much much greater influence than the stocks with the lowest prices. (PCLN is fetching $216, while CY only trades for $10.55.)

like the Dow Averages, many people will take issue in this basket being share price weighted. it's 'not fair' that some stocks have more influence simply because they cost more per share. the critique is valid, but whatever, just because the composition isn't fair doesn't mean the basket isn't useful.

now to the charts. first up is the basket over the past 616 trading days:

the broad sweeps look a lot like AAPL, don't they?

now let's turn the magnifier up a couple notches and view the past 351 days:

the chart formation since the March lows is clearly that of a five wave advance. this differs from most broader market indexes that look more like ABCs. the disconnect should not be surprising, since our basket's express purpose is to represent the purely bullish essence of an unpure cyclical rally. it is when this basket completes the ideal bull market formation that we anticipate the reemergence of the primary bear market in the general market.

so if our wave count is right -- and as I say, it looks pretty clear-cut to me -- then we are in the final leg of the overall advance. the big question is whether this presumed fifth wave has hit its peak or not.

could be, but if so this last wave covered a lot fewer points than waves one and three. the first two impulse waves advanced at least 200 points, while the latest one has only managed about 100 points so far.

I'll be on the lookout for signs of a top, but honestly I don't see much evidence of one yet. in any case, our goal is not to call the top in this basket, but for the basket to help inform us as to the viability of the market at large (basis your preferred index).

the performance of the 'Leaders of the Pack' basket will be updated each day in the 'Index Data' posts. I also plan on keeping tabs on the basket in the newsletter on an as needed basis. so you'll need to sign up for the pay service (top of the blog) if you want to follow me following this.