+
−% Training session Topic 5 exercise 04
+
−% 
+
−% Fitting time series with polynimials
+
−% 
+
−% 1) Load time series noise
+
−% 2) Fit data with ao/polyfit
+
−% 3) Check results
+
−% 
+
−% 
+
−% L FERRAIOLI 22-02-09
+
−%
+
−% $Id: TrainigSession_T5_Ex04.m,v 1.2 2011/05/13 15:13:12 ingo Exp $
+
−%
+
−%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
+
−
+
−%% 1) load fsdata
+
−
+
−% load test noise AO from file
+
−a = ao(plist('filename', 'topic5/T5_Ex04_TestNoise.xml'));
+
−a.setName;
+
−
+
−% looking at data
+
−iplot(a)
+
−
+
−%% 2) polyfit
+
−
+
−% ao/polyfit is a wrapper of the matlab function polyfit
+
−
+
−% set plist for fit
+
−plfit = plist('N', 6); % set the order of the polynomial to be fitted to data
+
−
+
−% fit data
+
−p = polyfit(a, plfit);
+
−
+
−% evaluating model
+
−b = ao(plist('polyval', p, 't', a));
+
−
+
−%% 3) Check results
+
−
+
−% plot model and data
+
−iplot(a,b)
+
−
+
−% plotting fit residuals
+
−iplot(a-b)
+
−
+
−
+
−
+
−
+
−