| boxplotCsub(x,g,notch,sym,vert,whis,c,fillit,LineWidth,koutline,cur_tot,fatter,lessoverlap,meansym)
|
function [h,h_ticklabels] = boxplotCsub(x,g,notch,sym,vert,whis,c,fillit,LineWidth,koutline,cur_tot,fatter,lessoverlap,meansym)
%BOXPLOTCSUB Display boxplots of a data sample, allowing line property definitions
% Modification of standard BOXPLOT from MatLab 2007a
% See below for old BOXPLOT syntax
%
% BOXPLOTCSUB calls BOXUTILC to do the actual plotting.
%
% [H,H_TICKLABELS] = BOXPLOTCSUB(X,G,NOTCH,SYM,VERT,WHIS,C,FILLIT,LINEWIDTH,KOUTLINE,[CUR TOT],FATTER,LESSOVERLAP,MEANSYM)
% INPUT
% X Data (see below for format)
% G OPTION : produces a box plot for the vector X grouped by G (optional, see below)
% NOTCH TRUE to include notches (default is FALSE).
% SYM Symbol and color to use for all outliers (default is 'r+').
% VERT vertical Box orientation (default is TRUE)
% WHIS Maximum whisker length (default 1.5). Value 0 omits whiskers.
% C OPTION : define color of box/line/sym
% FILLIT OPTION : TRUE plots filled boxes (default is FALSE).
% LINEWIDTH OPTION : set LineWidth, 0.5 default
% KOUTLINE OPTION : if true uses 'k-' for the line color while c option us still
% used for the fill. This allows for the outline of the box and
% the mean line to remain black and visible (default is FALSE).
% [CUR TOT] Subplot style boxplot allowing for multiple boxplots
% in the style of grouped bar plots. Easiest implementation using
% subplot style of calling with CUR bieng current plot out of TOT plots.
% no error checking yet to make sure TOT does not change between calls etc.
% FATTER OPTION : multiplier to increase the width of all boxes (Default = 1).
% LESSOVERLAP OPTION : reduce width of boxes to reduce overlap, in units (Default = 0.01).
% MEANSYM OPTION : marker symbol for showing mean value is used (Default is empty)
%
% OUTPUT
% H = BOXPLOT(...) returns the handle H to the lines in the box plot.
% H has one column per box, consisting of the handles for the various
% parts of the box.
%
% H Rows correspond to
% 1: upper whisker line
% 2: lower whisker line
% 3: lower extent endline
% 4: upper extent endline
% 5: box
% 6: median line
% 7: outlier marker
% 8: boxfill (if used)
% H_TICKLABELS graphics handle output to ticklabels
%
%BOXPLOT Display boxplots of a data sample.
% BOXPLOT(X) produces a box and whisker plot with one box for each column
% of X. The boxes have lines at the lower quartile, median, and upper
% quartile values. The whiskers are lines extending from each end of the
% boxes to show the extent of the rest of the data. Outliers are data
% with values beyond the ends of the whiskers.
%
% BOXPLOT(X,G) produces a box and whisker plot for the vector X grouped
% by G. G is a grouping variable defined as a categorical variable,
% vector, string matrix, or cell array of strings. G can also be a cell
% array of several grouping variables (such as {G1 G2 G3}) to group the
% values in X by each unique combination of grouping variable values.
%
% In a notched box plot the notches represent a robust estimate of the
% uncertainty about the medians for box-to-box comparison. Boxes whose
% notches do not overlap indicate that the medians of the two groups
% differ at the 5% significance level. Whiskers extend from the box
% out to the most extreme data value within WHIS*IQR, where WHIS is the
% value of the 'whisker' parameter and IQR is the interquartile range
% of the sample.
%
% Example : 2 sets divided into 3 groups of data
% mydataset = randn(10,2,3)
% boxplotCsub( mydataset(:,:,1),1,[],0,0,'r',true,0.5,false,[1 3])
% boxplotCsub( mydataset(:,:,2),1,[],0,0,'g',true,0.5,false,[2 3])
% boxplotCsub( mydataset(:,:,3),1,[],0,0,'b',1,0.5,false,[3 3])
% axis;axis([min(min(min(mydataset))) max(max(max(mydataset))) ans(3) ans(4)])
%
% Example : 2 sets divided into 3 groups of data (showing different format options)
% mydataset = randn(10,2,3);
% b_h1 = boxplotCsub( mydataset(:,:,1),1,['x'],1,1,'r',true,0.5,false,[1 3],1.5,0.005,false); % solid red, no mean symbol
% b_h2 = boxplotCsub( mydataset(:,:,2),1,['+'],1,1,'g',false,0.5,false,[2 3],1.5,0.005,false); % outline green, default mean symbol
% b_h3 = boxplotCsub( mydataset(:,:,3),1,['o'],1,1,'b',1,2,true,[3 3],3,0,'wd'); % solid blue, outline black, fatter, whitediamond mean symbol
% axis;axis([ans(1) ans(2) floor(min(min(min(mydataset)))) ceil(max(max(max(mydataset))))])
% legend([b_h1(8,1) b_h2(5,1) b_h3(8,1)],{'set1','set2','set3'},'Location','EastOutside')
%
% Example : 1 single data set with 4 columns, using boxplotCsub to colorize
% each box seperatly while keeping outline and median line black
% mydataset = randn(10,4);
% b_h = boxplotCsub( mydataset(:,:),1,['x'],1,1,'r',true,0.5,true,[1 1],1.5,0.005,false);
% set(b_h([1:4 7],4),'Color',color4);set(b_h(8,4),'FaceColor',color4)
% set(b_h([1:4 7],3),'Color',color3);set(b_h(8,3),'FaceColor',color3)
% set(b_h([1:4 7],2),'Color',color2);set(b_h(8,2),'FaceColor',color2)
% set(b_h([1:4 7],1),'Color',color1);set(b_h(8,1),'FaceColor',color1)
%
% If there are no data outside the whisker, then, there is a dot at the
% bottom whisker, the dot color is the same as the whisker color. If
% a whisker falls inside the box, we choose not to draw it. To force
% it to be drawn at the right place, set whissw = 1.
% Copyright 1993-2000 The MathWorks, Inc.
% $Revision: 2.12 $ $Date: 2000/09/01 20:02:28 $
% Modifications
% BFGK 26-mars-2002
% remove "cla" to allow for hold on overplotting
% c option to define color of box/line/sym
% fillit option : 1 plots filled boxes, 0 is default (normal)
% LineWidth option to set LineWidth, 0.5 default
%
% BFGK 8-aout-03
% allow for empty SYM = [] to omit outlier plotting
% not compatible with WHIS = 1
% [cur tot] subplot style boxplot allowing for multiple boxplots
% in the style of grouped bar plots. Easiest implementation using
% subplot style of calling with CUR bieng current plot out of TOT plots.
% no error checking yet to make sure TOT does not change between calls etc.
% BFGK 21-aout-03
% koutline if true uses 'k-' for the line color while c option us still
% used for the fill. This allows for the outline of the box and
% the mean line to remain black and visible. FALSE by default
% BFGK 17-oct-03
% h graphics handle output, outputs matrix of graphics
% handles, each column represents a data set. Rows
% correspond to
% 1: upper whisker line
% 2: lower whisker line
% 3: lower extent endline
% 4: upper extent endline
% 5: box
% 6: median line
% 7: outlier marker
% 8: boxfill (if used)
% BFGK 6-mai-2004
% included "mgrp2idx" as an internal function as it is private for
% boxplot in dealing with grouping variables
% BFGK 7-mar-2005
% corrected bug in strange case where there is only 1 datavec
% BFGK 10-jun-2010
% fatter multiplier to increase the width of all boxes. Default = 1
% lessoverlap reduce width of boxes to reduce overlap, in units. Default = 0.01
% BFGK 30-sep-2010
% added "h_ticklabels" output handle to tick label text when group variable is used. The
% labeltext when using g replaces the XTickLabel with text objects,
% so that there can be multiline labels, which are not possible with
% XTickLabel. The Position of these can then be modified afterwards.
% Note, with subplots, the label is generated for each subplot, so
% there are mutliple copies. This effect has not be modified as
% subplot can be used to shift boxplots for use with other plots,
% other than another boxplot...
% BFGK 1-oct-2010
% meansym marker symbol for showing mean value is used
% h(9,:)
if (nargin==1 & length(x(:))==1 & ishandle(x)), resizefcn(x); return; end
if nargin < 10,
error('boxplotCsub : insufficient number of arguments')
return
end
whissw = 0; % don't plot whisker inside the box.
[m n] = size(x);
if min(m,n) > 1
xx = x(:,1);
yy = xx;
else
n = 1;
xx = x;
yy = x;
end
% If the 2nd arg is not a grouping variable, shift arguments
nargs = nargin;
if (nargin<2)
g = [];
elseif (nargin>1 & (isequal(g,1) | isequal(g,0)))
if (nargin>12), meansym = lessoverlap ; end
if (nargin>11), lessoverlap = fatter ; end
if (nargin>10), fatter = cur_tot ; end
if (nargin>9), cur_tot = koutline; end
if (nargin>8), koutline = LineWidth; end
if (nargin>7), LineWidth = fillit; end
if (nargin>6), fillit = c; end
if (nargin>5), c = whis; end
if (nargin>4), whis = vert; end
if (nargin>3), vert = sym; end
if (nargin>2), sym = notch; end
notch = g;
g = [];
else
nargs = nargin - 1;
end
if (nargs < 2 | isempty(notch)), notch = 0; end
if (nargs < 3 ), sym = '+'; end
%if (nargs < 3 | isempty(sym)), sym = '+'; end
if (nargs < 4 | isempty(vert)), vert = 1; end
if (nargs < 5 | isempty(whis)), whis = 1.5; end
if (nargs < 6 | isempty(c)), c = ''; end
if (nargs < 7 | isempty(fillit)), fillit = 0; end
if (nargs < 8 | isempty(LineWidth)), LineWidth = 0.5; end
if (nargs < 9 | isempty(koutline)), koutline = 0; end
if (nargs < 11 | isempty(fatter)), fatter = 1; end
if (nargs < 12 | isempty(lessoverlap)), lessoverlap = 0.01; end
if (nargs < 13 | isempty(meansym)), meansym = []; end
if ~exist('cur_tot'),
error('boxplotCsub : insufficient number of arguments (cur_tot absent)')
return
end
if length(cur_tot) ~= 2,
error('boxplotCsub : cur_tot must have 2 elements ')
return
end
% Deal with grouping variable
if (~isempty(g))
x = x(:);
if (vert)
sep = '\n';
else
sep = ',';
end
[g,glabel,gname,multigroup] = mgrp2idx(g,size(x,1),sep);
n = size(gname,1);
k = (isnan(g) | isnan(x));
if (any(k))
x(k) = [];
g(k) = [];
end
end
% lf = n*min(0.15,0.5/n); % was just before '% Scale'
% lf = (n*min(0.15,0.5/n))/cur_tot(2);
lf = (n*min(0.15*fatter,fatter*0.5/n))/cur_tot(2);
%lb = 1:n;
sub_list = [1:cur_tot(2)] - mean([1:cur_tot(2)]) ;
lb = [1:n] + (lf*sub_list(cur_tot(1))) ;
xlims = [0.5 n + 0.5];
k = find(~isnan(x));
ymin = min(min(x(k)));
ymax = max(max(x(k)));
dy = (ymax-ymin)/20;
ylims = [(ymin-dy) (ymax+dy)];
% Scale axis for vertical or horizontal boxes.
% cla
set(gca,'NextPlot','add','Box','on');
if vert
axis([xlims ylims]);
set(gca,'XTick',[1:n]);
% set(gca,'XTick',lb);
set(gca,'YLabel',text(0,0,'Values'));
if (isempty(g)), set(gca,'XLabel',text(0,0,'Column Number')); end
else
axis([ylims xlims]);
set(gca,'YTick',[1:n]);
% set(gca,'YTick',lb);
set(gca,'XLabel',text(0,0,'Values'));
if (isempty(g)), set(gca,'YLabel',text(0,0,'Column Number')); end
end
set(gca,'UserData','boxplotCsub')
if (~isempty(g))
for i=1:n
z = x(g==i);
H(:,i) = boxutilC(z,notch,lb(i),lf,sym,vert,whis,whissw,c,fillit,LineWidth,koutline,lessoverlap,meansym);
end
if (multigroup & vert)
% Turn off tick labels and axis label
set(gca, 'XTickLabel','','UserData',size(gname,2));
xlabel('');
ylim = get(gca, 'YLim');
% Place multi-line text approximately where tick labels belong
for j=1:n
ht(j) = text(j,ylim(1),glabel{j,1},'HorizontalAlignment','center',...
'VerticalAlignment','top', 'UserData','xtick');
end
% Resize function will position text more accurately
set(gcf, 'ResizeFcn', sprintf('boxplot(%d)', gcf), ...
'Interruptible','off', 'PaperPositionMode','auto');
resizefcn(gcf);
elseif (vert)
set(gca, 'XTickLabel',glabel);
else
set(gca, 'YTickLabel',glabel);
end
elseif n==1
i=n; % added 7-mar-2005
vec = find(~isnan(yy));
if ~isempty(vec)
H(:,i) = boxutilC(yy(vec),notch,lb,lf,sym,vert,whis,whissw,c,fillit,LineWidth,koutline,lessoverlap,meansym);
end
else
for i=1:n
z = x(:,i);
vec = find(~isnan(z));
if ~isempty(vec)
H(:,i) = boxutilC(z(vec),notch,lb(i),lf,sym,vert,whis,whissw,c,fillit,LineWidth,koutline,lessoverlap,meansym);
end
end
end
set(gca,'NextPlot','replace');
if nargout > 0, h = H ; end
if nargout > 1, h_ticklabels = ht ; end
return
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
function resizefcn(f)
% Adjust figure layout to make sure labels remain visible
h = findobj(f, 'UserData','xtick');
if (isempty(h))
set(f, 'ResizeFcn', '');
return;
end
ax = get(f, 'CurrentAxes');
nlines = get(ax, 'UserData');
% Position the axes so that the fake X tick labels have room to display
set(ax, 'Units', 'characters');
p = get(ax, 'Position');
ptop = p(2) + p(4);
if (p(4) < nlines+1.5)
p(2) = ptop/2;
else
p(2) = nlines + 1;
end
p(4) = ptop - p(2);
set(ax, 'Position', p);
set(ax, 'Units', 'normalized');
% Position the labels at the proper place
xl = get(gca, 'XLabel');
set(xl, 'Units', 'data');
p = get(xl, 'Position');
ylim = get(gca, 'YLim');
p2 = (p(2)+ylim(1))/2;
for j=1:length(h)
p = get(h(j), 'Position') ;
p(2) = p2;
set(h(j), 'Position', p);
end
%%%% copied from private function for boxplot
function [ogroup,glabel,gname,multigroup] = mgrp2idx(group,rows,sep);
%MGRP2IDX Convert multiple grouping variables to index vector
% [OGROUP,GLABEL,GNAME,MULTIGROUP] = MGRP2IDX(GROUP,ROWS) takes
% the inputs GROUP, ROWS, and SEP. GROUP is a grouping variable (numeric
% vector, string matrix, or cell array of strings) or a cell array
% of grouping variables. ROWS is the number of observations.
% SEP is a separator for the grouping variable values.
%
% The output OGROUP is a vector of group indices. GLABEL is a cell
% array of group labels, each label consisting of the values of the
% various grouping variables separated by the characters in SEP.
% GNAME is a cell array containing one column per grouping variable
% and one row for each distinct combination of grouping variable
% values. MULTIGROUP is 1 if there are multiple grouping variables
% or 0 if there are not.
% Tom Lane, 12-17-99
% Copyright 1993-2002 The MathWorks, Inc.
% $Revision: 1.4 $ $Date: 2002/02/04 19:25:44 $
multigroup = (iscell(group) & size(group,1)==1);
if (~multigroup)
[ogroup,gname] = grp2idx(group);
glabel = gname;
else
% Group according to each distinct combination of grouping variables
ngrps = size(group,2);
grpmat = zeros(rows,ngrps);
namemat = cell(1,ngrps);
% Get integer codes and names for each grouping variable
for j=1:ngrps
[g,gn] = grp2idx(group{1,j});
grpmat(:,j) = g;
namemat{1,j} = gn;
end
% Find all unique combinations
[urows,ui,uj] = unique(grpmat,'rows');
% Create a cell array, one col for each grouping variable value
% and one row for each observation
ogroup = uj;
gname = cell(size(urows));
for j=1:ngrps
gn = namemat{1,j};
gname(:,j) = gn(urows(:,j));
end
% Create another cell array of multi-line texts to use as labels
glabel = cell(size(gname,1),1);
if (nargin > 2)
nl = sprintf(sep);
else
nl = sprintf('\n');
end
fmt = sprintf('%%s%s',nl);
lnl = length(fmt)-3; % one less than the length of nl
for j=1:length(glabel)
gn = sprintf(fmt, gname{j,:});
gn(end-lnl:end) = [];
glabel{j,1} = gn;
end
end
|
|
