Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Test for exponentiality

  1. Mar 19, 2013 #1
    If I have a sample of 150 data how to chose between tests of exponentiality?
    I mean is there any way to find the most appropriate test for my case judging on my data number,or maybe skewness, kurtosis, etc?
    Or there are 2 or 3 frequently used tests (Pearson maybe) which are most used?

    Please help
  2. jcsd
  3. Mar 22, 2013 #2

    Stephen Tashi

    User Avatar
    Science Advisor

    There is no mathematical answer to the "most appropriate" way until you define how to measure appropriateness. Even if you do that, there won't be be any mathematical answer unless you are willing to specify ( i.e. assume, guess, etc.) the exact way in which the data might be non-exponential.

    If you reveal the nature of the data, someone might know from experience what tests have worked well for them on data of that particular type. ("Worked well" can have various meanings too. It might mean "produced a publishable result", "pleased my thesis advisor" "was approved by a government agency", etc.)
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook