I've learned quite a bit after publishing this blog for almost three years so I've decided to reconfigure what I'm doing to hopefully be more useful. I've gone back and restated my results to take into account these changes. There are several reasons for this change: First, I've always considered earnings to be the key factor in stock market forecasting. Just recently I have been able to reconstruct the S & P earnings to exclude commodity companies and now I can actually project out earnings four quarters ahead with the data available at Portfolio 123. Secondly, I think I have been making the situation too complicated by looking at too many indicators and giving them too much weight in the model. Finally, I am looking for a way to spend less time on my investments and simplifying should be a big help.
For this week, my new earnings model, after going neutral for a brief time, has come back to a fully positive position. While valuation remains a serious problem, sentiment has been improving and is now classified as neutral. My model moves from 0% last week to 50% this week.
Earnings estimates of non-commodity companies in the S & P 500 are increasing on a forward 12 month basis. This sets my model at 100% exposure.
Sentiment factors have been improving as the market has consolidated for most of this year. Currently sentiment is neutral with no effect on over-all exposure.
Valuation remains extremely high and makes the market vulnerable. A discount of 50% is applied based on this overvaluation.
Total exposure this week = 100% -0% -50% = 50%
With my wife on Aruba