Asked by Locks
on 25 Mar 2013

I would like to have both:

a) a plot with the density function of a normal distributed random variable and the densitiy function of a log-normal distributed random variabel

b) a plot with the densitiy function of a normal distributed random variable and the density function of a t-distributed random variable for several degrees of freedom

*No products are associated with this question.*

Answer by Tom Lane
on 27 Mar 2013

Accepted answer

You could simply use the normpdf, lognpdf, and tpdf function to compute these densities, and plot the results together.

If you need to fit the distributions to data, here's an example that might get you started:

x = wblrnd(10,2,1000,1); [heights,centers] = hist(x,40); area = (centers(2)-centers(1))*sum(heights); bar(centers,heights/area,'hist') % normalize histogram to integrate to 1 p1 = fitdist(x,'normal'); % fit three distributions p2 = fitdist(x,'lognormal'); p3 = fitdist(x,'tlocationscale'); xx = linspace(0,30); line(xx,pdf(p1,xx),'color','r') % plot three densities line(xx,pdf(p2,xx),'color','g') line(xx,pdf(p3,xx),'color','m')

Locks
on 29 Mar 2013

Thanks!

I have changed to code to:

x = wblrnd(10,2,1000,1); [heights,centers] = hist(x,40); area = (centers(2)-centers(1))*sum(heights); bar(centers,heights/area,'hist') % normalize histogram to integrate to 1 p1 = fitdist(x,'normal'); % fit three distributions p2 = fitdist(x,'lognormal');

xx = linspace(0,30); line(xx,pdf(p1,xx),'color','r') % plot three densities line(xx,pdf(p2,xx),'color','g')

Because I only need the normal and the log normal distribution. Is there a way I can plot only those tho curves without the blue bars?

Tom Lane
on 1 Apr 2013

You can omit lines 2-4 (hist/area/bar) if you don't want the densities superimposed on a histogram.

Locks
on 1 Apr 2013

thanks!

Related Content

MATLAB and Simulink resources for Arduino, LEGO, and Raspberry Pi

Learn moreOpportunities for recent engineering grads.

Apply Today
## 0 Comments