What's the optimal look back time when evaluating equities? In some cases I get drastically different results by looking back 10 years as opposed to 5, etc. (07/09/2012)
Drawing Graphs: Shorter Time Frame vs Longer Time Frame
Although it is a very important question, there is no exact answer which would apply in all cases. However, there are some helpful suggestions that we can offer you. First and foremost, your goal should be to try to draw a graph that makes the most sense. In some cases this may mean removing an anomalous number that skews the results. For example, if a company you are drawing with a ten-year graph has nine years of similar growth like say 8% to 12%, but one year where earnings grew at 400%; this last number distorts the true average. Therefore, you would want to draw a graph for a time frame that eliminated the bad number.
Also, I find it useful to start out drawing a 20-year graph, and then reducing the number of years in increments of five. In other words, draw a 20-year then a 15-year followed by a 10-year followed by a 5-year. This exercise allows you to determine if the earnings growth rate is accelerating or decelerating. This is obviously important information to have. However, keep in mind that the F.A.S.T. Graphs™ tool is dynamic automatically calculates and applies the appropriate earnings multiples to various levels of growth (GDF, P/E=G, and GDF...P/E=G).