Economist Robert Shiller has developed a unique variation of the P/E ratio that uses the inflation-adjusted average from the previous 10 years earnings of the S&P 500 to calculate P/E ratios. What this does, according to Shiller, is smooth out the ". . . frequent boosts and declines that we see due to the business cycle" (Irrational Exuberance, 2000, p. 7).
In addition to the unique calculation methodology, it's also a fascinating set of market data that spans stock-market history from 1871 to present including two world wars, the bubble of the 1920s, the Great Depression, etc. The data set (if taken dirctly from Shiller's site) also includes columns for the consumer price index and 10-year interest rates.
Now that you know what it is and the theory behind it, this unique variation for calculating P/E ratios can be studied by either:
Equity markets, on average, go through periods of high and low valuations that can last for years. Although it is not impossible to find undervalued stocks during times of higher valuations, it is easier to find bargains during times of lower valuations. So, how do you tell when the market is over or undervalued?
The two methodologies below are the ones that I take seriously and consider from time to time. Each one offers a very-brief explanation as to why I they work. The link in the 'Total Market Cap to GNP' contains an essay written by Warren Buffet that may be helpful in understanding market cycles and historical periods of under- and over-valuation.