Curve Fitting and Over-optimization
A major stumbling block for many market timers is the phenomenon of curve fitting. This means that a given set of trading rules has been over-optimized to fit the particular dataset for which it has been back-tested. Unfortunately, if the trading rules are over-optimized they often fail to work on future data. Market timers attempt to avoid these difficulties in a number of ways. One is by looking for clusters of parameter values which work particularly well. Another is using out-of-sample data, which ostensibly allows the market timer to see how the system will work on unforeseen data. However, critics charge that once the strategy has been revised to reflect such data it is no longer "out-of-sample".
Read more about this topic: Market Timing
Famous quotes containing the words curve and/or fitting:
“In philosophical inquiry, the human spirit, imitating the movement of the stars, must follow a curve which brings it back to its point of departure. To conclude is to close a circle.”
—Charles Baudelaire (1821–1867)
“We do not quite say that the new is more valuable because it fits in; but its fitting in is a test of its value—a test, it is true, which can only be slowly and cautiously applied, for we are none of us infallible judges of conformity.”
—T.S. (Thomas Stearns)