AR prediction leads fitted data
1 次查看(过去 30 天)
显示 更早的评论
I am using aryule to construct a predictive AR filter. After I filter the data, the predicted values "lead" the actual data by one step. To illustrate this, I placed two images . The first shows the total actual and predicted data together (blue is actual, and orange is predicted. The second shows a zoomed in portion, so that the lead is apparent.
I don't understand where this lead comes from, nor do I know how to get rid of it in a justifiable way. Should I just shift the predicted data ahead by one step? Is this a relic of the filtering process? The code I'm using to get here is:
h = aryule(signal,26); %the 26 is the filter length
est_signal = filter(-h(2:end),1,signal);
0 个评论
回答(0 个)
另请参阅
类别
在 Help Center 和 File Exchange 中查找有关 Regression 的更多信息
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!