Take the 2-minute tour ×
Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. It's 100% free, no registration required.

I have a statistical maths problem where I have a rather large dataset consisting of a timestamp (rows) and a quantity of - lets say detections - per each day (columns for each location). Currently I have two challenges with the data:

Some of the data is missing, and I need to replace the missing values on some days or periods by a certain estimation based on earlier values. Is there a way to do this in Libreoffice/Matlab/Octave etc. tool, or what could be the best mathematical method to look into?

Second issue is that the data in a visualised form seems to follow a certain pattern, but the amount of detections is increasing gradually by time. Is it possible to average the whole dataset so that instead of increasing by time, the dataset would be more stagnant but still shows the same patterns periodically? This could be referred as to histogram equalization from image processing.

It would be great to have just possible statistical methods and/or tools to push to the right direction, so that I could do more research on the area and on the dataset on my own. :)

share|improve this question

1 Answer 1

up vote 1 down vote accepted

There are many ways to impute data. But most do not account for the added uncertainty due to the imputation. Multiple Imputation is a method that imputes each missing value several times and accounts for the uncertainty due to the imputation method. It is available in SAS and there is a special purpose software package for doing this that is called Solas which is licensed by Statistical Solutions. I don't know what is available in Matlab etc.

share|improve this answer
    
Thank you for this! I tested Multiple Imputation using R several times on my data, and looks like it works well in this case. –  tom.hj Jul 5 '12 at 12:31

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.