The R Project SVN R

Rev

Rev 59039 | Rev 65999 | Go to most recent revision | Show entire file | Ignore whitespace | Details | Blame | Last modification | View Log | RSS feed

Rev 59039 Rev 61433
Line 32... Line 32...
32 
  \code{n} is the number of data points.  If \code{span} is specified,
 32 
  \code{n} is the number of data points.  If \code{span} is specified,
33 
  a single smoother with span \code{span * n} is used.
 33 
  a single smoother with span \code{span * n} is used.
34 
 
 34 
 
35 
  The best of the three smoothers is chosen by cross-validation for each
 35 
  The best of the three smoothers is chosen by cross-validation for each
36 
  prediction. The best spans are then smoothed by a running lines
 36 
  prediction. The best spans are then smoothed by a running lines
37 
  smoother and the final prediction chosen by linear interpolation.
37 
  smoother and the final prediction chosen by linear interpolation.
38
38 
 
39 
  The FORTRAN code says: \dQuote{For small samples (\code{n < 40}) or if
 39 
  The FORTRAN code says: \dQuote{For small samples (\code{n < 40}) or if
40 
    there are substantial serial correlations between observations close
 40 
    there are substantial serial correlations between observations close
41 
    in x-value, then a pre-specified fixed span smoother (\code{span >
 41 
    in x-value, then a pre-specified fixed span smoother (\code{span >
42 
      0}) should be used.  Reasonable span values are 0.2 to 0.4.}
 42 
      0}) should be used.  Reasonable span values are 0.2 to 0.4.}
43 
 
 43