Friday, April 24, 2009

The New Bull Market In Banks?

This is probably the last chart I will show this week. I might put up one more tomorrow. But our two week chart fest is ending. Re all of the charts we have shown, this is a very useful nonproprietary visual I can share with you.


Markets tend to have an equilibrium point and linear regression analysis is helpful in determining where this equilibrium lies. A linear regression trendline is calculated to be a straight line that 'best fits' price between a starting point and an ending point in time. Error channels are calculated by plotting parallel lines above and below a linear regression trendline. In other words, a visual representation of mathematically calculated trendlines aka linear regression and its associated error channels.

Error analysis is helpful in determining where extremes lie as determined by error channels that are plotted a specified number of standard errors away from a linear regression trendline. Error channels can show if prices are cycling higher or lower than the linear regression trendline (equilibrium) and if a change in trend may be about to occur. Got that? It's simple. Even if you aren't a math whiz.

Semi-recently we highlighted the massive volume increase that has occurred in the XLF financial ETF over the last two years or so. Above is the linear regression trendline and its upper and lower error channels for the favorite ETF of traders since this mess began. The ETF holdings include most of the major screw ups on Wall Street.

Of course, the XLF is pretty doggone close to zero so we can't hold this channel forever. Does that mean we have hit a market bottom? No. And we will indirectly discuss one of the reasons why in a post on Germany over the next week or so.

Yappers telling us over and over again that the market has bottomed in the last two years have eaten enough turd sandwiches to last a lifetime. When a post mortem is written on this environment, most will claim that they called the bottom. What they won't tell you is that they called a dozen bottoms. These publicity hounds can't seem to stop themselves from impulsively looking more and more ridiculous. How often does someone have to be wrong before they eat a few doses of humble pie? Obviously when it's other people's money, it's an endless process. In other words, never.

As soon as the boy cries wolf (the market has bottomed) and no one listens, then we'll likely have hit a bottom.
posted by TimingLogic at 8:57 AM