+ Reply to Thread
Results 1 to 6 of 6

Thread: Pearson vs. Spearman: negative and positive values

  1. #1
    Points: 77, Level: 1
    Level completed: 54%, Points required for next Level: 23

    Posts
    4
    Thanks
    2
    Thanked 0 Times in 0 Posts

    Pearson vs. Spearman: negative and positive values




    Hey everyone, thanks for having and helping me .

    I calculated both Pearson and Spearman correlation coefficients on time series data of weekly returns of two assets (around 470 observations each). I got 0.017 for the Pearson correlation and -0.13977 for the Spearman. Is that realistic? In other words, is it possible to get such different values and even one being negative while the other is positive? Or did I miscalculate along the way?

    Many thanks for your help.

    Best regards,

    Mike

  2. #2
    TS Contributor
    Points: 17,779, Level: 84
    Level completed: 86%, Points required for next Level: 71
    Karabiner's Avatar
    Location
    FC Schalke 04, Germany
    Posts
    2,542
    Thanks
    56
    Thanked 640 Times in 602 Posts

    Re: Pearson vs. Spearman: negative and positive values

    Is that realistic?
    Yes.
    In other words, is it possible to get such different values
    Yes.
    and even one being negative while the other is positive?
    Yes. Change of sign is nothing sensational. One coefficient is just 0.15 lower than the other.

    Or did I miscalculate along the way?
    This possibility, of course, cannot be excluded.

    You already know the difference between Spearman rank coreelation and Pearson, I suppose.

    Have you already created a X-Y-scatterplot for your data?

    With Kind regards

    Karabiner

  3. The Following User Says Thank You to Karabiner For This Useful Post:

    MiUSA (10-31-2016)

  4. #3
    Points: 77, Level: 1
    Level completed: 54%, Points required for next Level: 23

    Posts
    4
    Thanks
    2
    Thanked 0 Times in 0 Posts

    Re: Pearson vs. Spearman: negative and positive values

    Thanks for your help, Karabiner. Yes, i plotted the returns of one assets against the other's. Plot is unspectacular an shows (as expected) no meaningful correlation (see plot). I know the difference, yes. for Pearson I simply calculated CORREL() in Excel, for Spearman, I ranked all values (with RANK.AVG) and then used CORREL() on these ranks. Many thanks!
    Attached Images  

  5. #4
    TS Contributor
    Points: 5,246, Level: 46
    Level completed: 48%, Points required for next Level: 104
    maartenbuis's Avatar
    Location
    Konstanz
    Posts
    372
    Thanks
    3
    Thanked 146 Times in 123 Posts

    Re: Pearson vs. Spearman: negative and positive values

    You have a number of outliers, and Spearman correlation is less sensitive to that. You can see what is happening by comparing the scatter plot you showed us with a similar scatter plot plotting the ranks agains one another.

  6. The Following User Says Thank You to maartenbuis For This Useful Post:

    MiUSA (10-31-2016)

  7. #5
    Points: 77, Level: 1
    Level completed: 54%, Points required for next Level: 23

    Posts
    4
    Thanks
    2
    Thanked 0 Times in 0 Posts

    Re: Pearson vs. Spearman: negative and positive values

    Thanks, maartenbuis. Indeed, the rank scatter plot looks like this and we might see a slight negative correlation, don't we?
    Attached Images  

  8. #6
    TS Contributor
    Points: 14,811, Level: 78
    Level completed: 91%, Points required for next Level: 39
    Miner's Avatar
    Location
    Greater Milwaukee area
    Posts
    1,171
    Thanks
    34
    Thanked 405 Times in 363 Posts

    Re: Pearson vs. Spearman: negative and positive values


    Need I say more? I think I see the Milky Way.
    Attached Images  

+ 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