vstd.m

JLAB is a set of Matlab functions I have written or co-written over the past fifteen years for the p

function[varargout] = vstd(varargin)
%VSTD Standard deviation over finite elements along a specfied dimension.
%
%   Y=VSTD(X,DIM) takes the standard deviation of all finite elements of X 
%   along dimension DIM.
%                                                                         
%   [Y,NUM]=VSTD(X,DIM) also outputs the number of good data points NUM, 
%   which is the same size as X.
%                                                                         
%   VSTD uses the "1/N" normalization, where N is the number of data points,
%   rather than the "1/(N-1)" normalization.
%
%   [Y1,Y2,...YN]=VSTD(X1,X2,...XN,DIM) also works.
%
%   VSTD(X1,X2,...XN,DIM);  with no arguments overwrites the original 
%   input variables.
%   __________________________________________________________________
%   This is part of JLAB --- type 'help jlab' for more information
%   (C) 2001, 2004 J.M. Lilly --- type 'help jlab_license' for details    
  
if strcmp(varargin{1}, '--t')
  vstd_test,return
end

dim=varargin{end};

for i=1:length(varargin)-1
  if all(isreal(varargin{i}))
     [varargout{i},numi{i}]=vmoment(varargin{i},2,dim);
  else
     %JML modified Sept 1 2004 to handle complex-valued data
     [u2,numi{i}]=vmoment(real(varargin{i}),2,dim);
     v2=vmoment(imag(varargin{i}),2,dim);
     varargout{i}=u2+v2;
  end
  varargout{i}=sqrt(varargout{i});
end

for i=length(varargin):nargout
  varargout{i}=numi{i-length(varargin)+1};
end

eval(to_overwrite(nargin-1))

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
function[]=vstd_test
x1=[1 2 3 nan];
x2=x1;
ans1=sqrt(2/3);
vstd(x1,x2,2);
reporttest('VSTD output overwrite', aresame(x1,ans1) && aresame(x2,ans1))
ans2=3;

x1=[1 2 3 nan];
[y1,y2]=vstd(x1,2);
reporttest('VSTD std & num', aresame(y1,ans1) && aresame(y2,ans2))

cv=randn(100,4)+sqrt(-1)*randn(100,4);
cvstd(:,1)=vstd(cv,1)';
cvstd(:,2)=vstd(real(cv),1)';
cvstd(:,3)=vstd(imag(cv),1)';
cvstd(:,4)=sqrt(cvstd(:,2).^2+cvstd(:,3).^2);
bool=aresame(cvstd(:,1),cvstd(:,4),1e-10);
reporttest('VSTD complex-valued data', bool)