For securities with underlying valuations that are difficult to calculate, the 200-session moving average is a good indicator of "underlying value."  Furthermore, this price represents a good technical trading equilibrium.  As I demonstrate with the S&P 500 ETF tracker, using this technique can give you better odds of timing the market to reduce your risk. 

    Since MSCI Index ETFs provide a good window into interest in the global market, I've decided to carry this analysis out on these ETFs as well.  Just as I'd done for SPY, I've calculated the percentage above or below the 200-session moving average at each time in the data set.  This gives an analog to premium/discount figures.  In this case, however, there are 21 resulting time series - one for EWA, EWC, EWD, EWG, EWH, EWI, EWJ, EWK, EWL, EWM, EWN, EWO, EWP, EWQ, EWS, EWT, EWU, EWW, EWY, EWZ, and EZU.  As a result, I've taken the average of this premium/discount percentage, giving a rough idea of the average premium/discount percentage for the global market. 

    This first graph is the premium/discount percentage up to Monday's high and yesterday's slight selloff.  As you can see when you expand the image, with respect to the MSCI ETFs, the global market has been extended at least 10% further in the past.  While this doesn't mean the market isn't overextended, there was a larger "bubble"  in early 2004 by this standard.

    This is the graph of the simultaneous premium/discount to the past 200 session average and the return over the following 200 sessions.  As you can see, there was a positive correlation between the two between 2001 and 2002, but in recent years, the correlation has become increasingly negative as liquidity has increased in the market.  Again, you'll want to click through to the expanded image.

    In conclusion, given the current negative relationship between the average premium/discount percentage and the subsequent return, I would hesitate right now before investing heavily in broad global funds.