MathOverflow is a question and answer site for professional mathematicians. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

I have a covariance matrix that I am trying to interpolate using kriging interpolation. The point of the kriging is to statistically predict any unknown point in-between know points. For example if I know the temperature at 9 locations that create a 3x3 square I can estimate the temp of any point inside the 3x3 square.

Each of the matrix values is a covariance value of the corresponding value on the 3x3 grid and the location that you want to find. I have been looking for a way to calculate covariance and can't seem to find any. The only information I was able to pull from the internet is that when using kriging, the geometry of the the data points are important in estimating the value you want because you need to know the influence of the point on your current location.

share|cite|improve this question
There is no question stated above! Also, perhaps stats.stackexchange will be a better venue for this question. – Suvrit Mar 1 '12 at 14:51
I personally would like to see questions on stuff like kriging here, but with such statistical terms of art explained for general context. I don't think, e.g., that stats.SE is much use for topological data analysis or nonlinear dimensionality reduction, both of which are statistical topics. – Steve Huntsman Mar 1 '12 at 18:14
@Steve: I have nothing against stats questions, but if you read the post, it's hard to figure out what really is the question! – Suvrit Mar 2 '12 at 2:47

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

Browse other questions tagged or ask your own question.