B=fillmissing(A,'nearest');%replace NaN with nearest non-NaN value
n nearest non NaN values
7 次查看(过去 30 天)
显示 更早的评论
I have a long time series (T= 150,000) with some NaN values in it, and I need to perform a regression using a rolling window.
That is, at each time t, I need to regress the past 60 non-NaN values on some other vectors of size (60,1). How can I extract the 60 non-NaN values closest (in the past) to t?
For example, if
A = [30 12 14 NaN 2 5 NaN 90 2 4 5 6 2 2 78 45 12 NaN 4];
and, for each t I need at least 4 past non-NaN values, the first regression I will perform will be at t=5, and will have Y = [30 12 14 2], the second regression will be at t=6, and will have Y = [12 14 2 5], and so on
0 个评论
回答(1 个)
另请参阅
类别
在 Help Center 和 File Exchange 中查找有关 Time Series 的更多信息
产品
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!