Mercurial > hg > octave-lyh

comparison scripts/polynomial/polyval.m @ 7501:83cce070104f

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
style fixes
author John W. Eaton <jwe@octave.org>
date Wed, 20 Feb 2008 00:46:30 -0500
parents 2df882e69f13
children 7cbe01c21986
comparison
equal deleted inserted replaced
7500:2df882e69f13 7501:83cce070104f
22 ## @deftypefnx {Function File} {@var{y}=} polyval (@var{p}, @var{x}, [], @var{mu}) 22 ## @deftypefnx {Function File} {@var{y}=} polyval (@var{p}, @var{x}, [], @var{mu})
23 ## Evaluate the polynomial at of the specified values for @var{x}. When @var{mu} 23 ## Evaluate the polynomial at of the specified values for @var{x}. When @var{mu}
24 ## is present evaluate the polynomial for (@var{x}-@var{mu}(1))/@var{mu}(2). 24 ## is present evaluate the polynomial for (@var{x}-@var{mu}(1))/@var{mu}(2).
25 ## If @var{x} is a vector or matrix, the polynomial is evaluated for each of 25 ## If @var{x} is a vector or matrix, the polynomial is evaluated for each of
26 ## the elements of @var{x}. 26 ## the elements of @var{x}.
27 ## @deftypefnx {Function File} {[@var{y}, @var{dy}] =} polyval (@var{p}, @var{x}, @var{S}) 27 ## @deftypefnx {Function File} {[@var{y}, @var{dy}] =} polyval (@var{p}, @var{x}, @var{s})
28 ## @deftypefnx {Function File} {[@var{y}, @var{dy}] =} polyval (@var{p}, @var{x}, @var{S}, @var{mu}) 28 ## @deftypefnx {Function File} {[@var{y}, @var{dy}] =} polyval (@var{p}, @var{x}, @var{s}, @var{mu})
29 ## In addition to evaluating the polynomial, the second output 29 ## In addition to evaluating the polynomial, the second output
30 ## represents the prediction interval, @var{y} +/- @var{dy}, which 30 ## represents the prediction interval, @var{y} +/- @var{dy}, which
31 ## contains at least 50% of the future predictions. To calculate the 31 ## contains at least 50% of the future predictions. To calculate the
32 ## prediction interval, the structured variable @var{s}, originating 32 ## prediction interval, the structured variable @var{s}, originating
33 ## form `polyfit', must be present. 33 ## form `polyfit', must be present.
73 A = (x(:) * ones (1, n+1)) .^ (ones (k, 1) * (n:-1:0)); 73 A = (x(:) * ones (1, n+1)) .^ (ones (k, 1) * (n:-1:0));
74 y(:) = A * p(:); 74 y(:) = A * p(:);
75 y = reshape (y, size (x)); 75 y = reshape (y, size (x));
76 76
77 if (nargout == 2) 77 if (nargout == 2)
78 ## The line below is *not* the result of a conceptual grasp of statistics.
79 ## Instead, after reading the links below and comparing to the output of Matlab's polyval.m,
80 ## http://www.mathworks.com/access/helpdesk/help/toolbox/stats/index.html?/access/helpdesk/help/toolbox/stats/finv.html
81 ## http://www.mathworks.com/access/helpdesk/help/toolbox/curvefit/index.html?/access/helpdesk/help/toolbox/curvefit/bq_5ka6-1_1.html
82 ## Note: the F-Distribution is generally considered to be single-sided. 78 ## Note: the F-Distribution is generally considered to be single-sided.
83 ## http://www.itl.nist.gov/div898/handbook/eda/section3/eda3673.htm 79 ## http://www.itl.nist.gov/div898/handbook/eda/section3/eda3673.htm
84 ## t = finv (1-alpha, s.df, s.df); 80 ## t = finv (1-alpha, s.df, s.df);
85 ## dy = t * sqrt (1 + sumsq (A/s.R, 2)) * s.normr / sqrt (s.df) 81 ## dy = t * sqrt (1 + sumsq (A/s.R, 2)) * s.normr / sqrt (s.df)
86 ## If my inference is correct, then t must equal 1 for polyval. 82 ## If my inference is correct, then t must equal 1 for polyval.