Why is my chi-squared distribution such a rubbish representation?Have I done something wrong?

I have data from 1994-2014 for births by mothers ages. I want to test the distribution for the Under20's mother ages. I have plotted a histogram with the information and tried to test the chi2pdf.

figure

x2=[];

[n,x2] = hist(x2,28); %Compute histogram

hbar2 = bar(x2, n/trapz(x2,n)); %Bar plot of histogram

hold on

%chi-distriution theoretical overlay
Ychi2 = chi2pdf(80000:1:150000,20); 

hchi2 = line(80000:1:150000,Ychi2,'Color','b','LineWidth', 2);

Answer 1

[Star Strider]

‘I want to test the distribution for the Under 20's mother ages.’

Test them for what?

Plot this first:

t = 1994:0.5:2004;
x2=[108111,112615,119719,132456,134946,137312,136012,135971,130784,127828,122145,121072,116622,110959,108844,107741,110722,113537,118589,125732,130744];

figure(1)
plot(t, x2)
grid

Then figure out what you want to do with your data.

Seems I wasn’t far off in my original suggestion to do the fft (link) and go from there.

[Star Strider]

The Fourier transform will give you the frequency content of your data. That may be irrelevant information in your situation.