|
POST
|
Also, if you're going to do kriging on the residuals of your MLR model, recalculate the model without using Lat as a covariate. Otherwise you'll be "double-counting" (for lack of a better phrase) the spatial location.
... View more
12-26-2012
10:25 AM
|
0
|
0
|
7969
|
|
POST
|
If you have a good MLR model already, I wouldn't try to use the covariates as cokriging variables. If you want to try it anyway, in the Geostatistical Wizard, when you choose kriging on the first page, you can enter up to four datasets. The first one you enter is the variable you will interpolate, and the three additional datasets will be used as cokriging variables. "Kriging" is often called "residual kriging," and there's a reason for this: you always perform kriging on the residuals of some model. This model can be almost anything, but "regression kriging," "kriging with external drift," "universal kriging," and "linear mixed model kriging" all generally refer to the simultaneous estimation of the covariate coefficients and the kriging parameters. However, you may find success with doing a sequential estimation: first calculate the coefficients using your MLR model. Then calculate the residuals, and perform Simple kriging on these residuals (you should use Simple kriging instead of Ordinary kriging because you know that the mean of the residuals is 0). Then add these interpolated residuals back into the MLR predictions. You'll lose some power because you are sequentially calculating parameters (rather than simultaneously estimating them), but you should still get defensible results. As for comparing the values in one point to the average of the neighboring points, the Semivariogram/Covariance cloud is probably the best way to visualize this, but the result is a graph rather than a single correlation coefficient. If you really need to calculate the correlation coefficient (and you're ok with ignoring spatial correlation in the analysis), we have a tool called Neighborhood Selection that selects the neighbors of an input (x,y) location (use the same neighborhood parameters that you used in kriging). It will probably take a lot of work, but I'm sure you can write a Python script that will do what you're trying to do. I've never personally done this, so I don't want to try to outline an algorithm, but I'm sure all the tools are there to accomplish this task.
... View more
12-26-2012
10:22 AM
|
0
|
0
|
7969
|
|
POST
|
Universal kriging is available in the Geostatistical Wizard. It's one of the six kriging types. The trend is calculated from polynomials of the (x,y) coordinates; it does not support covariates other than the spatial location. However, covariates can be used as cokriging variables.
... View more
12-26-2012
08:17 AM
|
0
|
0
|
7969
|
|
POST
|
You may just need to enable the extension. In ArcMap, go to Customize pulldown menu and select Extensions. Make sure Geostatistical Analyst is checked. If that isn't the problem, you'll need to contact Esri support.
... View more
12-26-2012
08:13 AM
|
0
|
0
|
745
|
|
POST
|
Yes, EBK will work with data with zeroes and negative values. If you need to use a transformation in EBK, you'll have to use Empirical because Log-Empirical won't work with zeroes.
... View more
12-20-2012
01:01 PM
|
0
|
0
|
799
|
|
POST
|
You're correct; it is the zeros that are causing the options to be filtered. We normally have Logarithm, Box-Cox, and Arcsin, but all three methods are not defined for negative values or zeros. If you are looking at transformations in preparation for kriging, we support Normal Score Transformations with simple kriging, and several of these transformations are applicable for negative/zero data values. Normal score transformations are performed within the Geostatistical Wizard. EDIT: I misspoke. Arcsin is defined for data between 0 and 1 (inclusive). So it can work with zero values, but if any data values are negative or greater than 1, the option will be filtered. Because the option is filtered for you, you must have some data values larger than 1.
... View more
12-20-2012
11:15 AM
|
0
|
0
|
799
|
|
POST
|
Trend removal options are not built into Areal Interpolation, but there are a few things you might try. First, you may be able to remove some trend by correcting for anisotropy. It's one of the options on the covariance curve screen of the Wizard. If that doesn't work, there is a second method that may work, but it is more statistically questionable. You may be able to represent the polygons as points and use Global or Local Polynomial interpolation to get a general trend. You can then subtract off this trend from your polygonal data and use areal interpolation on the new detrended data. After the interpolation, you can add back the trend that you removed. This methodology is hard to justify if you're using Rate (Binomial) or Event (Overdispersed Poisson) areal interpolation, but it can be reasonably justified for Average (Gaussian).
... View more
12-11-2012
12:11 PM
|
0
|
0
|
881
|
|
POST
|
Any of the three can be justified, but I prefer fixed distance band among those three. This is because fixed distance band is the conceptualization where it is most clear that the test statistic will converge to a normal distribution with increasing sample size. If you're getting the results you want/expect from fixed distance band, I would stick with that.
... View more
11-20-2012
07:06 AM
|
0
|
0
|
1843
|
|
POST
|
http://video.arcgis.com/watch/2034/using-areal-interpolation-to-perform-polygon_dash_to_dash_polygon-predictions This video visually takes you through the process of building an areal interpolation model and making predictions to new polygons. It also shows how to use areal interpolation to predict polygons with missing data. It uses the same data and model as our areal interpolation workflow help topic. http://resources.arcgis.com/en/help/main/10.1/index.html#//0031000000qm000000 Enjoy.
... View more
11-15-2012
12:46 PM
|
0
|
0
|
1530
|
|
POST
|
There are several reasons that a kriging surface can have holes like that. The most likely reason is that the searching neighborhood is too small. Regardless, we recommend using empirical Bayesian kriging (EBK). It almost always gives more accurate predictions than traditional kriging.
... View more
11-07-2012
05:06 AM
|
0
|
0
|
1246
|
|
POST
|
We currently do not support true space-time kriging. You can perform kriging on each time slice independently, but you can't use temporal correlation directly in the kriging model.
... View more
10-29-2012
01:41 PM
|
0
|
0
|
583
|
|
POST
|
Whether kriging will work really depends on your data and the question you're trying to answer. We generally use environmental data in examples because this kind of data most often meets the kriging assumptions, but there is really no restriction on the source of the data. If the data meets the kriging assumptions and it makes sense to interpolate, kriging should work.
... View more
10-18-2012
07:17 AM
|
0
|
0
|
954
|
|
POST
|
I'm not familiar with the Spatial Statistics tutorial. You'll probably have more luck on the Spatial Statistics Forum.
... View more
10-16-2012
08:11 AM
|
0
|
0
|
555
|
|
POST
|
Esri just posted the video for one of my kriging presentations at the user conference in July. It's a little over an hour, and it goes through all the basics of kriging in Geostatistical Analyst. If you want to learn a bit more about kriging and how it works, this video is a good place to start. http://video.esri.com/watch/1796/concepts-and-applications-of-kriging Enjoy.
... View more
10-11-2012
09:02 AM
|
0
|
1
|
2533
|
| Title | Kudos | Posted |
|---|---|---|
| 2 | 01-16-2025 04:52 AM | |
| 1 | 10-02-2024 06:45 AM | |
| 2 | 08-23-2024 09:18 AM | |
| 1 | 07-19-2024 07:09 AM | |
| 1 | 08-21-2012 09:47 AM |
| Online Status |
Offline
|
| Date Last Visited |
02-25-2026
06:39 PM
|