spm_ar_demo_coeff.m.svn-base

try the matlab scripts to do various computations

disp('Fitting AR model to AR-5 process data with known coefficients');

secs=5;
ns=128;
t=[1/ns:1/ns:secs];
N=length(t);
noise_var=1^2;
noise=sqrt(noise_var)*randn(1,N);
a_true=[-1.8517,1.3741,0.1421,-0.6852,0.3506];
y=filter(1,[1,a_true],noise);
y=y(1:N);
figure
plot(t,y);
xlabel('Seconds');
title('Sample of AR-5 process');

total_variance=std(y)^2;
signal_variance=total_variance-noise_var;
snr=sqrt(signal_variance)/sqrt(noise_var);
disp(sprintf('SNR=%1.3f',snr));

disp('For model order p=5');
ar=spm_ar (y,5);

disp(' ');
disp('True coefficients');
disp(a_true);

disp(' ');
disp('Estimated coefficients');
disp(ar.a_mean');


for p=1:10,
    disp(sprintf('Now fitting model with p=%d coefficients',p));
    ar=spm_ar (y,p,0);
    logev(p)=ar.fm;
end
logev=logev-min(logev);
figure
bar(logev);
ylabel('Log Evidence');
xlabel('Model order');