finding multivariate pdf of normally distributed data?

3 次查看（过去 30 天）
Hydro
Hydro 2014-10-1
评论： José-Luis 2014-10-2
Assume my data is multivariate normally distributed. how i can plot the pdf of my first variable? say I have data matrix below with four variables
A =
5 11 10 8
9 7 6 12
4 14 15 1
4 14 15 1
9 7 6 12
my mean is say
M=
4 14 15 1
my covariance matrix is
Cov=
16 2 3 13
5 11 10 8
9 7 6 12
4 14 15 1
I used mvnpdf(A(:,1),M(:,1),cov(:,1)). any help would be appreciated.

请先登录，再回答此问题。

回答（1 个）

SK
SK 2014-10-2
Do you want the histogram of the data. If so, use the hist() function.
If you want a theoretical normal marginal pdfs from the covariance and mean, use the diagonal terms from the covariance matrix as the variance for each variable. So for example, for the first variable:
x = 4-3*16 : 0.01 : 4+3*16;
y = pdf('normal', x, 4, 16);
plot(x(:), y(:));

类别

在 Help Center 和 File Exchange 中查找有关 Probability Distributions and Hypothesis Tests 的更多信息

标签

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by