Latest Contributions by EricKrause

The Coefficient values are the coefficients in the linear equation in the section titled "Regression analysis terms and concepts."  They're the B's in the equation. As for whether 10-30% error is acceptable, only you can answer that question.  It depends on your goals coming into this analysis. ... View more

The "Coefficient" values are the coefficients in the linear equation.  "Correlation coefficients" are something different; they measure the degree of correlation between the dependent variable and a single independent variable.  When you say that you "haven't received the value of dependent variable," do you mean that the linear equation doesn't perfectly predict the dependent variable?  If so, OLS does not make perfect predictions.  If your model is good, it should predict something close to the true value, but it won't be perfect. ... View more

Is it possible to make a layer package of your geostatistical layer and send it to ekrause@esri.com? ... View more

Unless you use barriers, all distances are currently calculated with Euclidean distance.  As you noted, this means that interpolating around the world is a bad idea. ... View more

This can happen if several of your independent variables are highly correlated.  This is called multicollinearity.  The variables distribution and relationships graph will give a better sense of how the independent variable is related to the dependent variable. ... View more

I just tested, and I get the same graphs whether I take the logarithm in the histogram tool or make a new field and calculate the logarithm.  I suspect you've made a mistake somewhere, but I don't know where it could be.  Try it again and see if it works this time. The Logarithm function in the histogram isn't doing anything special.  It is just taking the natural logarithm, exactly the same as the Log() function in the field calculator. ... View more

I'm not able to reproduce your issue.  Try restarting your computer and recreating the kriging layer.  If the problem persists, you'll need to contact Esri support services. ... View more

Kernel Interpolation can slow to a standstill if your barrier is too complicated.  You should be able to resolve this by running Simplify Polygon on your barrier to reduce the number of vertices.  That should get Kernel Interpolation working fast again. ... View more

The problem is probably that the bandwidth is too small.  You can only make a prediction at a location if there is an input point within the bandwidth distance.  You should also look into Kernel Interpolation with Barriers.  It generally outperforms Diffusion Interpolation with Barriers. ... View more

If you're still confused after reading that blog, try reading this forum thread: http://forums.arcgis.com/threads/77248-kriging-results-to-raster ... View more

The cell size is measured in map units.  To determine your map unit, hover over the map and look at the bottom right corner.  If it isn't in meters, you'll need to do a unit conversion (for example, if it is in kilometers, you'll want a cell size of .001).  If it is in Decimal Degrees (lat-long), you'll have to project your data because there is no direct unit conversion from decimal degrees to meters. ... View more

Not directly.  If you're using a constant inclusion probability raster, the algorithm will place points as far away from each other as possible.  So, the distance to the nearest neighbor will get smaller if you output more points.  That distance can get arbitrarily small if you output enough points. The only way to ensure that the points will be far apart is to keep running the tool with different numbers of points until you get the separation distance that you want.  If you're stuck to a particular number of points, there isn't much you can do. ... View more

You'll fit the cross-covariance curve the same way you fit a semivariogram.  You'll see empirical covariances (blue crosses), and you need to manipulate the parameters (range, nugget, sill, lag size, etc) to get the curve to go through the crosses as best you can.  The first thing to try should be the "Optimize Model" button.  Hopefully that will be able to fit everything automatically, though you may want to do some manual tweaking from there. Also, if you're doing external research, the cross-covariance curve is often called a "covariogram." ... View more

See my response here: http://forums.arcgis.com/threads/79627-joining-two-datasets-into-one ... View more