How can I find a fit for my data?

phi = [ones(size(x2)),x2,x2.^2,x2.^3,x2.^4];
aa = phi\y2;
yfit = phi*aa;
plot(x2,y2,'bx',x2,yfit,'k-')
grid on 
s=sprintf('y=%8f+%8fx+%8fx^2+%8fx^3+%8fx^4',aa(1),aa(2),aa(3),aa(4),aa(5));
legend('data',s);
cc = corrcoef(y2,yfit);

so I plotted the datas and is shown in blue crosses but the fit I found in black line is not accurate enough, how can I increase my fit accuracy?

Also I plotted data3 on a graph and notice its a trignometric function. But I dont know how to find a fit for this kind of graph.

Answer 1

[Cris LaPierre]

Consider exploring various fits using the curve fitting toolbox.

Also, there is a curve fitting app (if you have access to it).