Analyzing the used metrics
As mentioned the metrics used to evaluate a company are Earnings, Book Value, Sales and Dividends.
To find out if a given metric is useful to evaluate the future performance of a given stock, I will compare these metrics with the returns that where generated in the next few years to come. To achieve this I examined the correlation between the different metrics, and the returns in the next 5 years.
Because it is hard to do this for a large number of companies, analyze the data and conclude some results, instead of a lot of separate companies I used the S&P500 index as a single company.
If the valuation works well for a large index like the S&P500 it will work on average for separate companies as well.
The reason why I have chosen this index because it is big enough to give a good representation of the whole market, the other reason is that there is data available from 1871, and the third one is that this method may be used directly, if we just want to do index investing, to find the right entrance and exit points.
Because some of the data is available only from 2000, I will use two time horizons to analyze data: one starting from 1881 until 2015 and another starting from 2000 until 2015.
Here is a graph with the value of the S&P500 starting from 1881.
Based on this graph it is quite clear why to invest in the stock market. Starting from less than 5$ it reached more than 2000 in today, so the stock market is a great investment, however if we look at the graph below, it is clear that we need to choose the entry and exit points well.
The returns for the S&P500 for the shorter and longer period are below.
So the purpose of this analysis is to find some kind of correlation between Earnings, Book Value, Sales, Dividends, and the bars shown in the graphs below.
Based on the metrics mentioned above, I will try to find an algorithm to find out if the market is overvalued or undervalued. If such a formula can be found, the same formula can be applied than to separate companies as well.