The most useful EMA indicator for trading gold over nearly 32 years was to buy gold when the U.S. dollar crosses below its 7-week EMA and to sell gold when the dollar crosses above its 7-week EMA. This inverse strategy would have gained more than 889% before trading costs and without any leverage. Nearly 52% of the 282 trades would have been profitable, which is pretty good for a trend-following strategy.
Gold moves directly with the trend of the Swiss franc (using a 5-week EMA crossover) and the trend of crude oil (using a 12-week EMA crossover). These make good intuitive sense: Gold should go up when the Swiss franc (a hard currency with gold backing) and crude oil (another inflation-sensitive commodity) are rising.
My testing found an inverse relationship with gold and 10-year T-note yields. This might be spurious, however, because it lacks logic and has been unprofitable since 2003, as 10-year interest rates and gold both rose together. I also doubt the validity of trading gold based on the DJIA or S&P 500. Stocks are not always a good hedge against inflation, and this indicator's Cumulative Equity Curve is too erratic to be reliable.
I conclude that the U.S. dollar (inverse trend), Swiss franc (direct trend), and crude oil (direct trend) are the best and most logical external relationship trend indicators for gold. ...........................


These indicators have been much more profitable than the trend of the gold price itself.
Robert W. Colby is an analyst/strategist for TradingEducation.com and a consultant to institutional and private investors and traders. A serious student of investing for nearly 40 years, Colby is the author of The Encyclopedia of Technical Market Indicators. (TradingEducation.com) (robertwcolby.com)

http://www.marketwatch.com/news/story/divining-golds-next-move/story.aspx?guid={5CABD9A0-0CAC-4439-88F0-1A3E23A9F412}