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 am trying to calculate the change frequency for a set of data. Each bit of data has the date-time it was created. I would like to say for a specific set of data the change frequency is hourly, daily, weekly, monthly or yearly.

So far I have tried getting the list of dates and get the min/max which is easy to calculate an average from which can be converted into a human readable label such as hourly, daily etc

How would i take into account the age of the last new bit of data. eg: say there were 50 dates all roughly an hour one after the other. This is hourly. but if the last one was 2 weeks ago, its not quite hourly.

I am sure there is a formula to calculate this, but I don't know where to start.

Thanks

P.S. I took a guess at the tags, not sure if there are better ones to use for this question.

share|improve this question
    
Sounds subjective. You're the only one who knows what you really want. If you're not going to be more precise than that in the formulation of the problem, we can't really help you other than producing a (potentially endless) list of suggestions. For instance, in the one case you mention, what do you want to classify it as? Weekly? Hourly anyway? Daily? Who's gonna tell if YOU don't know? –  Raskolnikov Jul 6 '11 at 20:12
    
hi, I am not to sure what it should be. If I knew I would not be asking for any help :P. Just thought I would get some direction here as anything searched on google with frequency in it brings up light and sound waves. –  dogmatic69 Jul 6 '11 at 20:15
    
I'm afraid we don't know either then. It's not a math problem, you see. It's a modelisation problem. It all depends on what you want. Once you know what you want, then the math kicks in, in that we can tell you what mathematical tools will help you realize it. As such, I think your question would fit better on some other SE. Maybe programmers? Or cross validated? –  Raskolnikov Jul 6 '11 at 20:17
    
Anyway, I will hazard a guess and I think what you need is some statistical tools for the analysis of clustering in time series. Which would fit well on Cross Validated. –  Raskolnikov Jul 6 '11 at 20:21
    
thanks, I have registered on there. will see what they got to say about it :) –  dogmatic69 Jul 6 '11 at 20:24

Your Answer

 
discard

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

Browse other questions tagged or ask your own question.