How does that joint gaussian distribution in the beggining and the conditional expectation of one variable based on the other relate to kriging estimation?
The equations that PEST uses to estimate parameters can be re-formulated as conditioning equations that pertain to a Gaussian distribution. For PEST, one conditions k (parameters) by h (observations). When kriging one conditions k at one place by k at other places. In the first case covariances between h and k are calculated by the model (assumed to be linear). In the second case covariances are embodied in C(k), the covariance matrix between parameters that is derived from a variogram. I hope that this (at least partially) answers your question. Best wishes John
![Seungmin "그렇게, 천천히, 우리(As we are)" | [Stray Kids : SKZ-PLAYER]](http://i.ytimg.com/vi/kAzmhLHePqU/mqdefault.jpg)
Thanks a lot. You made it understandable
spectacular!
How does that joint gaussian distribution in the beggining and the conditional expectation of one variable based on the other relate to kriging estimation?
The equations that PEST uses to estimate parameters can be re-formulated as conditioning equations that pertain to a Gaussian distribution. For PEST, one conditions k (parameters) by h (observations). When kriging one conditions k at one place by k at other places. In the first case covariances between h and k are calculated by the model (assumed to be linear). In the second case covariances are embodied in C(k), the covariance matrix between parameters that is derived from a variogram.
I hope that this (at least partially) answers your question.
Best wishes
John