I use two market indices and two time frames to determine overall direction and strength of the U.S. stock market. The Russell 3000 index is cap-weighted index that represents the largest 3000 companies in the U.S. It tracks very closely to the S&P500 index. The Value Line Arithmetic Index is an equal weighted index of approximately 1,650 stocks. The VLA provides an estimate of how an equal-dollar weighted portfolio of stocks will perform. Using a combination of both indices, I can get a pretty good idea of the trend the U.S. equity market.
I use a 200 day moving average as a benchmark to measure the long term direction of the market. Although my model was designed to indicate intermediate term risk, I’ve learned it’s a good idea to use a longer periods to provide macro context and weight to shorter term indicators. The 75 day moving average is a reasonable measure of the intermediate term direction of the market. It serves as the early warning that a trend change may be in the making.
The score of each trend indicator is determined by the daily close of the index and the five day moving average compared to its benchmark moving average. The five day smoothing minimizes whipsaws that can produce jarring changes model scores.
Each trend indicator can produce one or three possible scores: +1, -1, or zero. Below is an example of how I score each trend indicator.
+1 = close and 5 dma > 200 dma
0 = close > 200 dma, 5 dma < 200 dma
0 = close < 200 dma, 5 dma > 200 dma
-1 = close and 5 dma < 200 dma
Friday, November 30, 2007
Subscribe to:
Post Comments (Atom)
1 comments:
In my graduate school days in the behavioral sciences, a category that could just as easily labeled inexact sciences, "construct validity" was a popular term. In essence it means that even though the underlying theory is hard to empirically improve beyond doubt, does it "look" right and meet the rigors of analytical reasoning. And, what you are sharing sure does look good. At some point down the road, I'd be interested in hearing your own critique. In what kind of mkt environments is it working best, which ones maybe not as good. And, of course, nothing is perfect and as you indicate the purpose of this is to smoothe the ride rather than time every high and low.
Post a Comment