I have two time-based data sources (one providing data by month and the other by week) that must be combined to create a third daily source (recognizing that at best we're getting one possible projection of those two data sources). To make things really interesting, they do not cover entirely the same timeframe (the monthly stops at the most recent month end at least 45 days ago, while the weekly data covers only the last six weeks). The main issue (for me) is addressing the days where the two datasources overlap.

Not looking for someone to give me a fish, but for some papers on how to fish (actually, I hate fishing).

  • 1
    $\begingroup$ Do they overlap at all? 45 days > 6 weeks! $\endgroup$ – Jeffε Aug 6 '12 at 22:16

(not enough rep for comment) Can you tell us more about the data streams? I am a roboticist, and I often deal with sensors which are outputting data streams at different rates. Specifically, can you please elaborate on the following:

  • do the data streams have the same modality? (eg, are they all telling you the velocity of something, or is one telling you velocity and the other position, leaving you to infer velocity?)
  • is either stream noisy and/or can you otherwise get conflicting data from the streams? If so, do you have an error model?

Once I know more about your streams, I will try to suggest some relevant literature.

  • $\begingroup$ These datastreams represent sales data from a distributor in two different forms. Both streams represent number of items sold over a certain time period (week or month). Theoretically there should be no conflicts; in practice, I have seen conflicts between them. Conflict handling is part of what I'm looking for help with. But primarily $\endgroup$ – Jack R-G Aug 20 '12 at 23:47
  • $\begingroup$ The crux of what I'm looking for is an algorithm that would create a projection of these two datasets into a non-conflicting third dataset containing daily sales (while there are likely to be multiple non-conflicting solutions, I only need one). $\endgroup$ – Jack R-G Aug 20 '12 at 23:54

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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