📄 kern_dens_plot.m
字号:
function h=kern_dens_plot(returns,bw,nodes,range,kern)
% PURPOSE:
% Plots a bivariate kernel smooth
%
% USAGE:
% h=kern_dens_plot(returns,bw,nodes,range,kern)
%
% INPUTS:
% returns : T by 2 length series to generae the data
% bw : The bandwidth to use. If empty, silverman's BW is used
% nodes : Number of nodes to use
% range : Range to calculate the nodes over(2 by 1 vector)(can be empty, in which case the range is used)
% kern : String arguement of the type of kernel to be used
% 'cosinus'
% 'epanechnikov'
% 'normal'
% 'quartic'
% 'triangular'
% 'triweight'
% 'uniform'
%
% OUTPUTS:
% h : Graphics handle to plos
%
% COMMENTS:
% Uses product kernels
%
% Author: Kevin Sheppard
% kevin.sheppard@economics.ox.ac.uk
% Revision: 2 Date: 12/31/2001
if isempty(bw)
bw=1.06*std(returns)*length(returns)^(-1/5);
end
if isempty(range)
rmin=min(returns);
rmax=max(returns);
else
rmin=range(1);
rmax=range(2);
end
if isempty(nodes)
nodes=length(returns);
end
height=zeros(nodes,1);
steps=linspace(rmin,rmax,nodes);
index=1;
for i=steps
height(index)=(1/bw)*mean(feval(kern,returns,steps(index),bw));
index=index+1;
end
h=plot(steps,height);
⌨️ 快捷键说明
复制代码
Ctrl + C
搜索代码
Ctrl + F
全屏模式
F11
切换主题
Ctrl + Shift + D
显示快捷键
?
增大字号
Ctrl + =
减小字号
Ctrl + -