+ Reply to Thread
Results 1 to 3 of 3

Thread: Regression? or Time series analysis?

  1. #1
    Points: 7, Level: 1
    Level completed: 13%, Points required for next Level: 43

    Posts
    1
    Thanks
    0
    Thanked 0 Times in 0 Posts

    Regression? or Time series analysis?




    Hi, this is my first time posting.

    I am trying to prove a point- but it's been a while since I've dabled in statistics.

    My null hipotesis is "It never snows on christmas". And "If it snows on christmas it won't on New Years day". I have the data for 30-50 years weather it snowed or not on the day (actually how much it snowed but I can recode that). I don't want to just say it is 75% chance it will snow or in 75% it did snow, but I would like a statisticaly accurate result.

    So I am thinking for first hypothesis I need anova, for the second regression? Is that right?
    I will go back and reread my notes from school, but would like some direction.

    I remember we also did something with time series analysis, does someone have a short explanation what it is and if it could come in handy with such data and similar hipotheses?

    Thank you for the help,
    Oliva

  2. #2
    Points: 24, Level: 1
    Level completed: 47%, Points required for next Level: 26

    Posts
    3
    Thanks
    0
    Thanked 0 Times in 0 Posts

    Re: Regression? or Time series analysis?

    Frankly I don't know that either ANOVA or any time series method is really appropriate here.

    The simplest procedure given only these data would seem to be estimating the conditional probabilities:

    P(\text{it snows on New Year's Day} | \text{it snows on Christmas})

    and:

    P(\text{it snows on New Year's Day} | \text{it doesn't snow on Christmas})

    By doing no more than taking note of all the years in which it snowed on Christmas and then taking the fraction of those years where it went on to snow on New Year's Day and then doing the same thing with every year where it did not snow on Christmas, then choosing which probability to go with on the coming Christmas according to whether it snowed or not on that day.

    Of course the weather forecast going towards New Year's Day is likely an even better estimate.

  3. #3
    Fortran must die
    Points: 58,790, Level: 100
    Level completed: 0%, Points required for next Level: 0
    noetsi's Avatar
    Posts
    6,532
    Thanks
    692
    Thanked 915 Times in 874 Posts

    Re: Regression? or Time series analysis?


    I agree that time series is not very useful here. You can use Christmas to predict snow in ANOVA or regression and see if that is statistically significant. Where time series might come into play is if there is a trend in the data so that snow on Christmas changes over time. You can do a chow test to detect this.
    "Very few theories have been abandoned because they were found to be invalid on the basis of empirical evidence...." Spanos, 1995

+ 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