+ Reply to Thread
Results 1 to 2 of 2

Thread: Sample size | Johnson SB distribution

  1. #1
    Points: 24, Level: 1
    Level completed: 47%, Points required for next Level: 26

    Posts
    1
    Thanks
    0
    Thanked 0 Times in 0 Posts

    Sample size | Johnson SB distribution




    I have some statistical notions but from one day to the other i was asked to do an analysis that i hope you guys can help me with, since i have no idea how i will be able to do it.

    I have data from the last 5 years of a chemical company. This company receives a material from their suppliers and to test if it's acceptable or not they do a density test on this material. They receive it several times a day and they test every single delivery so I have lots of data (from the results of the density test). I have put this data into excel and it seems that it follows a johnson Sb distribution, negatively skewed.

    What they asked me to do was, "since there are lots of costs associated with the density test we want to determine the sample size that would give us 95% confidence that average of the test results will be between x and y (error)."
    If the distribution was normal i think it would be easy but since it is negatively skewed i'm kind of stuck on this. What steps should i take? Can I turn it into a Normal distribution?

    Thank you very much

  2. #2
    Devorador de queso
    Points: 95,819, Level: 100
    Level completed: 0%, Points required for next Level: 0
    Awards:
    Posting AwardCommunity AwardDiscussion EnderFrequent Poster
    Dason's Avatar
    Location
    Tampa, FL
    Posts
    12,935
    Thanks
    307
    Thanked 2,629 Times in 2,245 Posts

    Re: Sample size | Johnson SB distribution


    Once you take a large enough sample the distribution of the sample mean will be approximately normal. So you could do the calculations the same way you would if you assumed the distribution was normal and check the sample size you get from that to see if it seems large enough for the central limit theorem to kick in.

    Alternatively you could just bootstrap the sampling distribution of the sample mean for various values of n until you find one that meets your requirement.
    I don't have emotions and sometimes that makes me very sad.

+ Reply to Thread

           




Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts






Advertise on Talk Stats