Download as pptx, pdf, or txt
Download as pptx, pdf, or txt
You are on page 1of 8

Probability Plot Examples

Jake Blanchard
Spring 2008

Uncertainty Analysis for Engineers 1


Making Your Own Prob Plots
 Sort data
 Formulate CDF
 Linearize distribution for CDF
 Fit to straight line
 Plot both

Uncertainty Analysis for Engineers 2


An Example (7.4 from text)
 Exponential Distribution

f   exp   x  a 
F  1  exp   x  a 
1  F  exp   x  a 
ln 1  F    x  a 
 ln 1  F    x  a    x   a

 Slope is 
Uncertainty Analysis for Engineers 3
Code
data=[blah blah];
data=sort(data);
n=numel(data);
for i=1:n
plotcdf(i)=-log(1-i/n);
end
data=data(1:n-1);
plotcdf=plotcdf(1:n-1);
p=polyfit(data,plotcdf,1);
f=polyval(p,data);
plot(data,plotcdf,'x',data,f)
slope=p(1)
invslope=1/slope Uncertainty Analysis for Engineers 4
Plot
4

3.5

2.5

1.5

0.5

0
0 500 1000 1500 2000 2500 3000

Uncertainty Analysis for Engineers 5


Example 7.5
 Gumbel Extreme Value

F  exp  exp    y  u 


ln( F )   exp    y  u 
 ln  ln( F )    y  u    y   u

 Slope=
 Intercept= u

Uncertainty Analysis for Engineers 6


Code
data=[blah blah];
n=numel(data);
for i=1:n
plotcdf(i)=-log(-log(i/n));
end
data=data(1:n-1);
plotcdf=plotcdf(1:n-1);
p=polyfit(data,plotcdf,1);
f=polyval(p,data);
plot(data,plotcdf,'x',data,f)
slope=p(1) Uncertainty Analysis for Engineers 7
Plot
4

-1

-2
4.5 5 5.5 6 6.5 7 7.5

Uncertainty Analysis for Engineers 8

You might also like