Re: Combining histograms and function plots

[Miquel Cabanas]

hi,

On Wed, Jan 29, 2003 at 03:45:47PM +0100, Toke Reichstein wrote:
>
> Q1: How do I plot a normal distribution function with the mean and
> standard deviation of my data in Octave?

use the function "normal_pdf()". I'm not sure it's part of the
standard distribution, if you fail to find it go and download it
from the sourceforge repository

http://octave.sourceforge.net

> Q2: How do I plot it ontop of my histogram plots?

see the example below.


# Since i've no real data i'll fake some normally distributed
# random one (matrix size 100*100)
data = randn (100);

# you can check it with. Note that i use data(:) so that it is
# treated as a one-row vector instead of as a matrix.
plot (data(:))

# now let's calculate some statistics. Again, i use data(:) to
# treat it as a single-row vector. Try "mean (data)" to see the
# difference.
mean_data = mean (data(:));
var_data = var (data(:));

# now hist() can be used to plot the histogram. Hist() has several
# options, check them in the online help. The way it's written
# below will make an histogram of data(:), with bins from -5 to 5
# and as many bins as given by the increment: 101 bins, and it will
# be normalized to area "1" (last argument).
hist (data(:), [-5:.1:5], 1)

# now hold the plot so that the next plot doesn't overwrite it
hold on

# it's time to use normal_pdf to calculate our distribution probability.
# For this i give the x-values, the mean value of the distribution,
# and the variance
data_pdf = normal_pdf ([-5:.1:5], mean_data, var_data);

# then i normalize data_pdf to area one
data_pdf = data_pdf ./ sum (data_pdf) ;

# now I can plot the distribution stored in data_pdf
plot ([-5:.1:5], data_pdf)


Also check bar() if you plan to manipulate your bins.


Miquel


-- 
Miquel E Cabanas ------------------------------------------------------
SeRMN, Universitat Autonoma de Barcelona (address@hidden)
MRUI Software Manager (address@hidden)
------------------------------------------o-oo--ooo---ooo--oo-o--------



-------------------------------------------------------------
Octave is freely available under the terms of the GNU GPL.

Octave's home on the web:  http://www.octave.org
How to fund new projects:  http://www.octave.org/funding.html
Subscription information:  http://www.octave.org/archive.html
-------------------------------------------------------------