3

I am by no means a math expert, but I am analyzing very large weather data data for a computer science course .I am taking away all the weather related issues out of the analyzing and just looking strictly at it in the eyes of statistics.

Lets say there are 2 events, 1 occurs 250 times in the year, and the other event occurs 50 times in a year. The second event occurs every single day the first event occurred. Can I say that there is a relationship between the two events? I am not sure if it is safe to say there is a good chance there is a relationship between the data because this could just be a coincidence. For example if the event A occurred 25 times a year and event B occurred 15 times again event B occurring every time event A occurred I would be more confident to say there is a relationship between the two events because there is less of a chance it is a coincidence due the the number of times event A occurred. To rephrase, lets say event A occurred 365 times in the year and event B occurred 10 times in the year, event B would occur every single time event A occurred, but this is just because event A occurred every day so there is no relationship. Is there some sort of standard to say that when two events have a statistically high probability of having a relationship of some sort?

Hope this makes sense.

user2924127
  • 221
  • 1
  • 6
  • How many years of data do you have? – onetoinfinity Jan 07 '15 at 22:35
  • Currently I am working on a small scale of 50 days worth of data while developing and testing, but will have a dataset of 11-13 years worth of data. – user2924127 Jan 07 '15 at 22:37
  • Once you get you hands on the larger piece of data, obviously check it and see if your relationship still holds. If it does, then the probability of that happening randomly over a decade seems unlikely enough that you can claim a relationship. However, be sure not to falsely associate causation, as there may be another phenomenon behind both event 1 and 2. – onetoinfinity Jan 07 '15 at 22:41
  • The dataset will check for relationships in the short term and longterm (check to see if there is a relationship recently say in the last 3 months, 1 year, but also longer 4 years, 8 years, 10 years). For shorter datasets is there any methods? – user2924127 Jan 07 '15 at 22:44

0 Answers0