+ Reply to Thread
Page 2 of 2 FirstFirst 1 2
Results 16 to 19 of 19

Thread: testing rates through time

  1. #16
    Points: 855, Level: 15
    Level completed: 55%, Points required for next Level: 45
    d21e7x11's Avatar
    Location
    Hamilton, ON Canada
    Posts
    62
    Thanks
    7
    Thanked 9 Times in 9 Posts

    Re: testing rates through time




    shanehall.m, I added to the code/comments/output in my posts - sorry for the confusion.
    Note that the interaction can only be tested in the models where time is continuous.
    Last edited by d21e7x11; 10-12-2011 at 12:57 PM.

  2. #17
    Points: 1,935, Level: 26
    Level completed: 35%, Points required for next Level: 65

    Location
    USA
    Posts
    29
    Thanks
    10
    Thanked 0 Times in 0 Posts

    Re: testing rates through time

    So from what I understand, first I must apply the arcsine transformation to the proportions. Then fit each drug to a regression line and run an anova model testing to see if the drugs are different. I do not know R code too well... Is year a repeated measure variable?

  3. #18
    Points: 1,935, Level: 26
    Level completed: 35%, Points required for next Level: 65

    Location
    USA
    Posts
    29
    Thanks
    10
    Thanked 0 Times in 0 Posts

    Re: testing rates through time

    I see now... thank you... In our case we probably want the transformed data and time as continous. So to see if there was a difference in the trends of the drugs, we would be looking at the interaction, correct? What that output is saying is that there is no difference in the trends or in the drugs, but there is a difference in the rates over the years. We probably need more than 4 years of data to get reliable estimates for the terms, especially interaction and time. Is my thinking correct?

  4. #19
    Points: 1,935, Level: 26
    Level completed: 35%, Points required for next Level: 65

    Location
    USA
    Posts
    29
    Thanks
    10
    Thanked 0 Times in 0 Posts

    Re: testing rates through time


    drug A would have a sample size > 200,000
    Drub b would have a sample size > 7,000
    drug c would have a sample size > 600

    That is as accurate as I can get.
    Does this change anything?

+ Reply to Thread
Page 2 of 2 FirstFirst 1 2

           




Tags for this 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