Skip to main content
12 events
when toggle format what by license comment
Apr 13, 2017 at 12:44 history edited CommunityBot
replaced http://stats.stackexchange.com/ with https://stats.stackexchange.com/
Jun 2, 2016 at 18:40 comment added Steffen Moritz If you have a multivariate time series you can apply Amelia. It does not work for univariate time series, because it's algortihms employs inter-attribute correlations to impute the data. This is not possible in the univariate case, because there is just one attribute.
Jun 2, 2016 at 18:32 comment added Steffen Moritz As you can see in this link (arxiv.org/ftp/arxiv/papers/1510/1510.03924.pdf) Amelia does not work for univariate time series.
Nov 26, 2015 at 8:19 comment added MRI @Dawny33 The missing data is not due to human or any other error.That is reason I want to separate.Or apply methods directly without filling the missing data.
Nov 26, 2015 at 7:33 comment added user81847 I was hoping a self-contained answer. Forget.
Nov 26, 2015 at 7:30 comment added Dawny33 @Pascal That is clearly explained in the documentation, with a clear example
Nov 26, 2015 at 7:28 comment added user81847 For example, you create a time serie with missing value and you show how to implement the Amelia solution.
Nov 26, 2015 at 7:27 comment added Dawny33 @MRI Why would you want to separate the data based on the breaks. If the missing data is not due to human or any other error, then doing so makes else. Else, No
Nov 26, 2015 at 7:26 comment added Dawny33 @Pascal What do you mean by a concrete example? These links are enough for the OP to handle the problem.
Nov 26, 2015 at 7:23 comment added MRI Thank You.Is it possible to separate the data based on the breaks abd apply some methods ?
Nov 26, 2015 at 7:22 comment added user81847 Please show a concrete example, not only links.
Nov 26, 2015 at 7:18 history answered Dawny33 CC BY-SA 3.0