There is a risk to this sort of long-term analysis - they identify 'following the trend' as a good strategy; but it was a good strategy under different circumstances. The advent of computers is a huge deal that may have changed the game.
Even as recently as 1970, how easy was it to execute a trend following strategy? Gathering and processing the data would have been a mammoth activity requiring several clerks and lots of record keeping. Demonstrating the it was a good strategy would also have been very difficult (more clerks, lots of by-hand calculation). Only a serious trading house could have pulled it off. The ability of actuaries to employ statistical models has also changed completely in the last 10-20 years. The amount of money following a trend-following strategy is also an extremely interesting but unknowable figure.
If the advent of computers makes it trivial even for a roaming programmer to implement a trend following strategy, is it still going to be robust? Now that it can be demonstrated to be a strong strategy, will the market correct for it?
Even as recently as 1970, how easy was it to execute a trend following strategy? Gathering and processing the data would have been a mammoth activity requiring several clerks and lots of record keeping. Demonstrating the it was a good strategy would also have been very difficult (more clerks, lots of by-hand calculation). Only a serious trading house could have pulled it off. The ability of actuaries to employ statistical models has also changed completely in the last 10-20 years. The amount of money following a trend-following strategy is also an extremely interesting but unknowable figure.
If the advent of computers makes it trivial even for a roaming programmer to implement a trend following strategy, is it still going to be robust? Now that it can be demonstrated to be a strong strategy, will the market correct for it?