[peterboro31]

http://www.moneysense.ca/2010/02/09/predict-the-next-crash/


For those who wish to view this. The only problem is the lag with the ma's; but I setup a chart for TSX composite, and it certainly shows that it crosses over in upswings and downswings.

What's your opinion of this?


http://ca.moneycentral.msn.com/inve...CF=8&D7=&D6=&symbol=$CA:OSPTX&nocookie=1&SZ=0

 

[andrewf]

There's plenty of evidence that when markets are below their moving average (can use 200 day, but the result is not very sensitive to length of the moving average), there is an increase in volatility. It does not necessarily mean that there will be a crash. A timing system can be useful in reducing risk, but it won't necessarily improve returns.

Mebane Faber wrote a fairly accessible paper on using moving averages in investing. I've mentioned it a number of times before, but here is the think:

A Quantitative Approach to Tactical Asset Allocation

As he notes, moving averages are hardly a new idea, but despite the fact that the market is aware of the strategy, it still seems to work. He theorizes that this is a result of cognitive biases of investors, such as loss aversion, and thus not readily changed.