Results 1 to 9 of 9

Math Help - Comparing Gaussian curves

  1. #1
    Newbie
    Joined
    Mar 2011
    Posts
    6

    Comparing Gaussian curves

    Hi,

    I have two points A(30.333,100) and B(30.333151665,80)


    I need to draw Gaussian for each point separately and need to check their overlap area for comparison.

    Since these two the distance between these two curves is 5 ppm (when compared with x - axis values). I would like to draw the Gaussian curves on ppm scale as base line.

    I need to solve this problem. How can I calculate "meu" and sigma for such kind of comparisons and what is the exact formula which gives the distance between these two 2D graphs
    Follow Math Help Forum on Facebook and Google+

  2. #2
    Grand Panjandrum
    Joined
    Nov 2005
    From
    someplace
    Posts
    14,972
    Thanks
    4
    Quote Originally Posted by asd1231 View Post
    Hi,

    I have two points A(30.333,100) and B(30.333151665,80)


    I need to draw Gaussian for each point separately and need to check their overlap area for comparison.

    Since these two the distance between these two curves is 5 ppm (when compared with x - axis values). I would like to draw the Gaussian curves on ppm scale as base line.

    I need to solve this problem. How can I calculate "meu" and sigma for such kind of comparisons and what is the exact formula which gives the distance between these two 2D graphs
    Please post the entire question, or the context.

    CB
    Follow Math Help Forum on Facebook and Google+

  3. #3
    Newbie
    Joined
    Mar 2011
    Posts
    6
    Hi,
    Thanks for the reply.

    Actually I have two points with x and y coordinates and I want to compare these two points. So, that I will get a score which says how good they are.

    What I want to do is to draw a gaussian curve against each point and overlap them and check for the overlapping area.

    Is it correct what I'm doing?
    Follow Math Help Forum on Facebook and Google+

  4. #4
    Senior Member
    Joined
    Oct 2009
    Posts
    340
    So, you are assuming that points A and B come from separate multivariate normal distributions? You can't estimate sigma (sigma being a matrix) for the two distributions without more observations.

    Also, what exactly do you mean by "how good they are?"
    Follow Math Help Forum on Facebook and Google+

  5. #5
    Newbie
    Joined
    Mar 2011
    Posts
    6
    Hi,

    Yes, something like that.

    How good they are in the sense how good the fit is when you compare or overlap them by drawing Gaussian curves.

    Can you please tell me in details how could I calculate the meu and sigma and the formula for calculating the overlap area between two Gaussian.

    I'm sorry pleas don't mistake me for asking everything, this is very important for me. I don't want to assume any wrong calculations since these calculations now are very important in the biology field where I'm working in.
    Follow Math Help Forum on Facebook and Google+

  6. #6
    Grand Panjandrum
    Joined
    Nov 2005
    From
    someplace
    Posts
    14,972
    Thanks
    4
    Quote Originally Posted by asd1231 View Post
    Hi,

    Yes, something like that.

    How good they are in the sense how good the fit is when you compare or overlap them by drawing Gaussian curves.

    Can you please tell me in details how could I calculate the meu and sigma and the formula for calculating the overlap area between two Gaussian.

    I'm sorry pleas don't mistake me for asking everything, this is very important for me. I don't want to assume any wrong calculations since these calculations now are very important in the biology field where I'm working in.
    I'm afraid your posts read like you have no idea what you are doing or asking. That is why I want you to post the real question and/or more context.

    CB
    Follow Math Help Forum on Facebook and Google+

  7. #7
    Newbie
    Joined
    Mar 2011
    Posts
    6
    Hi,

    For example if you have a peak at a position of 30.33 on x - axis and has a height of 100. Can't we draw a gaussian against that peak?

    If we can then we draw peak at at a position of 30.337 on x - axis and has a height of 100.


    Now overlapping this two Gaussian curve and calculating the overlap area will give the fit.

    If they overlap perfectly then the fit is 1.

    I hope I make it clear this time
    Sorry, I know it is just a basic question but still it is a puzzle for me.
    Follow Math Help Forum on Facebook and Google+

  8. #8
    Senior Member
    Joined
    Oct 2009
    Posts
    340
    I agree with Captain Black. I still have only the vaguest notion of what you are trying to do (it looks to me like you are asking how to calculate the distance between two points in the plane using an unusual metric). Assuming that I even understand that correctly, I have no idea why anyone would want to do that (I know what Kullback-Liebler divergence is, but I don't see its usefulness in this context unless I'm missing something).

    I suggest you provide a bit more context or a more thorough explanation of what you are doing, or else talk to a statistician at your university. I actually think it would be helpful it you could draw a picture of what you are trying to do and posting it, as well.
    Follow Math Help Forum on Facebook and Google+

  9. #9
    Grand Panjandrum
    Joined
    Nov 2005
    From
    someplace
    Posts
    14,972
    Thanks
    4
    Quote Originally Posted by asd1231 View Post
    Hi,

    For example if you have a peak at a position of 30.33 on x - axis and has a height of 100. Can't we draw a gaussian against that peak?
    With the information you have supplies: No

    Or rather: yes, but it would be meaningless.
    Follow Math Help Forum on Facebook and Google+

Similar Math Help Forum Discussions

  1. Comparing two proportions
    Posted in the Statistics Forum
    Replies: 0
    Last Post: May 10th 2011, 12:04 PM
  2. Difference between Gaussian noise and white Gaussian noise?
    Posted in the Advanced Statistics Forum
    Replies: 2
    Last Post: February 12th 2011, 01:25 AM
  3. comparing mean
    Posted in the Advanced Statistics Forum
    Replies: 0
    Last Post: November 13th 2008, 01:40 PM
  4. Replies: 2
    Last Post: March 25th 2008, 08:38 AM
  5. sketching curves-transformation curves f(x)!!!!
    Posted in the Pre-Calculus Forum
    Replies: 3
    Last Post: December 2nd 2006, 07:55 AM

Search Tags


/mathhelpforum @mathhelpforum