Suppose for NA values in a perticular variable in a time series dataset what will be the best alternative to avoid NA values?
Taking the mean value of that perticular variable and replace NA or putting zero in place of NA values.
How to deal with missing values depends a lot on your situation. What you're trying to answer with your data, the process by which data is missing, and other factors are important to how you might want to deal with those NAs.
A good place to start thinking about this is Rob J Hyndman and George Athanasopoulos' book chapter, " Forecasting: Principles and Practice", "Dealing with missing values and outliers"
1 Like
I would just add that there is a 3rd edition Sec. 13.9 to the book that deploys the tidyverts tools that are easier to work with for this class of problem, imho.
This topic was automatically closed 21 days after the last reply. New replies are no longer allowed.
If you have a query related to it or one of the replies, start a new topic and refer back with a link.