Editing Extrapolation (section)

==Quality==

Typically, the quality of a particular method of extrapolation is limited by the assumptions about the function made by the method.  If the method assumes the data are smooth, then a non-[[smooth function]] will be poorly extrapolated.

In terms of complex time series, some experts have discovered that extrapolation is more accurate when performed through the decomposition of causal forces.<ref>{{cite journal|url= https://repository.upenn.edu/bitstreams/cbec1013-046d-453f-b6e6-eae86e3d3ccc/download | title = Decomposition by Causal Forces: A Procedure for Forecasting Complex Time Series |author1=J. Scott Armstrong |author2=Fred Collopy |author3=J. Thomas Yokum | journal = International Journal of Forecasting | year = 2004| volume = 21 | pages = 25–36 | doi = 10.1016/j.ijforecast.2004.05.001 | s2cid = 8816023 }}</ref>

Even for proper assumptions about the function, the extrapolation can diverge severely from the function.  The classic example is truncated [[power series]] representations of sin(''x'') and related [[trigonometric function]]s.  For instance, taking only data from near the ''x''&nbsp;=&nbsp;0, we may estimate that the function behaves as sin(''x'')&nbsp;~&nbsp;''x''.  In the neighborhood of ''x''&nbsp;=&nbsp;0, this is an excellent estimate.  Away from ''x''&nbsp;=&nbsp;0 however, the extrapolation moves arbitrarily away from the ''x''-axis while sin(''x'') remains in the [[interval (mathematics)|interval]] [&minus;1,{{nbsp}}1].  I.e., the error increases without bound.

Taking more terms in the power series of sin(''x'') around ''x''&nbsp;=&nbsp;0 will produce better agreement over a larger interval near ''x''&nbsp;=&nbsp;0, but will produce extrapolations that eventually diverge away from the ''x''-axis even faster than the linear approximation.

This divergence is a specific property of extrapolation methods and is only circumvented when the functional forms assumed by the extrapolation method (inadvertently or intentionally due to additional information) accurately represent the nature of the function being extrapolated.  For particular problems, this additional information may be available, but in the general case, it is impossible to satisfy all possible function behaviors with a workably small set of potential behavior.