Latest Contributions by LaurenRosenshein

This is a great question that we get a lot!  At this time logistic regression is not available in ArcGIS, but we do have a sample script available that helps you run logistic regression using the R statistical package right from inside ArcMap.  You can find the sample script here. ... View more

Hi Oliver, Great question!  Actually, we've got an awesome sample script for ArcGIS 10 that does exactly what you're asking for.  It is called Exploratory Regression and it tests all combinations of your explanatory variables for all of the assumptions of OLS (including spatial autocorrelation of the residuals).  You can find it here: Supplementary Spatial Statistics Toolbox.  Also, Exploratory Regression is going to be part of core starting with ArcGIS 10.1, so if you'd like to try the latest version of the tool you might want to consider joining the Beta Program. ... View more

Hi Stephanie, If you want to learn more about Hot Spot Analysis and all of the Spatial Statistics tools, our resources page: http://esriurl.com/spatialstats has tons of short videos, tutorials, sample scripts, and lots of other stuff that should help you get started! ... View more

Hi Gina, Thanks so much for being willing to share your data with us so that we can figure this out!  My email address is [email protected].  Please send over your data, as well as a screenshot of the parameters you're choosing, and any additional information that might be useful.  I look forward to solving this problem and getting you moving forward! Lauren Rosenshein Geoprocessing Product Engineer ... View more

Hi Antonio, To start with, it definitely sounds like you're on the right track by using the Spatial Autocorrelation tool to figure out a good distance to use for your Cluster and Outlier Analysis.  Actually, there is a sample script in the resource center that does this analysis for you (and has a lot of documentation about how to use this method), which is called Incremental Spatial Autocorrelation, in the Supplementary Spatial Statistics Toolbox.  The output that you're seeing looks alright (without having seen your data).  It isn't uncommon to see p-values that are all 0.00000.  It means that at every distance you've tried you're getting statistically significant clustering.  The key is to look for peaks in the z-scores associated with those statistically significant p-values which occur at a specific distance.  Peaks in the z-score represent distances at which the spatial processes promoting clustering are the most pronounced.  If you download the Supplementary Spatial Statistics Toolbox there is a folder called Documentation, which has a document that helps you Learn More about Incremental Spatial Autocorrelation.  You can also find tons of other resources about the spatial statistics tools here:  http://esriurl.com/spatialstats. ... View more

Hi Jeanee, This is a great question, and one that we get quite a bit. With GWR, there is a local linear equation for each feature in the dataset. The equation is weighted so that nearby features have a larger influence on the prediction of yi than features that are farther away. I do know that our consultant�??s GWR software (Fotheringham, Charlton and Martin) does compute p-values for each coefficient in every one of the local linear equations. However, because doing so is really not appropriate (and we�??ve discussed this with our consultants and they agree with that assessment), we do not report the coefficient p-values for our ArcGIS GWR tool. I know our consultants are looking into other methods for computing those p-values that might be appropriate, but I don�??t believe they have come up with an ideal solution. Because GWR does not have the strong diagnostics (like p-values, as just one example) that OLS does, we very strongly recommend finding a properly specified OLS model before moving to GWR. Unless you are only interested in predictions (not interested in the coefficients�?� in variable relationships), you cannot trust GWR results unless you can be sure you�??ve found all of the key explanatory variables to model your dependent variable (you can see this easily: run OLS or GWR with 1 important explanatory variable and examine the coefficients; add a second important explanatory variable and notice that the coefficient values change�?� sometimes dramatically; the coefficient values can change 180 degrees, in fact). Having said that, we have just released a tool and documentation on the Geoprocessing Resource Center to help you find a properly specified OLS model: Exploratory Regression and �??What they don�??t tell you about regression analysis�?�. The Exploratory Regression tool is similar to Stepwise Regression except instead of just looking for high Adj R2 values, it looks for models that meet all of the assumptions of the OLS method (no variable redundancy, no spatial autocorrelation in regression residuals, no model bias, statistically significant coefficients�?�). To download this tool and the associated documentation, check out http://esriurl.com/spatialstats (look for Supplementary Spatial Statistics). I took this text right out of this forum post, which also has some additional useful information about using condition numbers and local r-squared results of GWR to see how well you're doing, as well as some other information about moving to GWR from OLS. ... View more

Hi Stephen, Sorry that you've run into this issue!  Actually, this sounds like a known issue that is on our list of bugs to fix.  Basically, what's happening is that we are doing all of the mathematical calculations using double precision, and that is what's used to create the actual geometry of the output feature.  The bug is that we are creating the xcoord field and the ycoord field as "Floats" with only single precision, which is why you're seeing differences when you run the Add XY Coordinates tool.  So, the true geometry of the feature is correct, but we are missing precision in the fields where we write the coordinates.  If you don't mind, we'd like to see if a simple code change will fix this issue for you.  You'll have to find the MeanCenter.py file, which will be located here: C:\Program Files (x86)(**Or wherever you have ArcGIS installed)**\ArcGIS\Desktop10.1\ArcToolbox\Scripts.  1.  The first thing to do is rename that file to MeanCenter_Orig.py (this way you'll have the original saved somewhere just in case) 2.  Then, open up the MeanCenter_Orig.py, and do a "Save As" and name it MeanCenter.py (the file has to have this name to be found by the Mean Center tool, so you'll want to make sure the file you update has this name) 3.  Once you've done this, do a search for the words "Add Field Names".  In this section of the code you will see something that looks like this: for fieldName in dataFieldNames:         UTILS.addEmptyField(outputFC, fieldName, "FLOAT") 4.  What you want to do is change the word "FLOAT" to "DOUBLE".  5.  Save your changes. Once you've made the change the xcoord and ycoord fields should be created as doubles, which means that they should have the right precision and you should no longer see any difference from when you run the Add XY Coordinates tool.  If you are still seeing differences, please let us know so that we can figure out what's going on! Thanks so much for bringing this to our attention, and again I'm very sorry that you've run into this! Lauren Rosenshein Geoprocessing Product Engineer ... View more

Hi gbateson, I'm really sorry that you're running into this problem too!  We'd really like to reproduce the issue so that we can figure out what's going on.  Would you be willing to share your data with us so that we can figure this out?  I'll also need a breakdown of exactly how you're filling out the tool parameters (a screenshot would work). Again, I'm very sorry that you're having trouble and I hope we can get it figured out soon! Lauren Rosenshein Geoprocessing Product Engineer ... View more

Hi Stephanie, There are tons of resources that will help you get started with your hot spot analysis.  You can find links to a lot of our resources here:  http://esriurl.com/spatialstats.  A good place to start might be several of the short videos, including Performing Proper Density Analysis, Hot Spot Analysis Part 1, Hot Spot Analysis Part 2, Hot Spot Analysis Part 3, and a one-hour free seminar called Introduction to Spatial Pattern Analysis.  There is also a Hot Spot Analysis Tutorial which I think you'll find very useful.  It walks step by step through performing a hot spot analysis, and helps you understand the decisions that you're making along the way. ... View more

Hi Ling Xue, There is actually a tool to do this in the Data Management Toolbox called Table to Ellipse.  It should help you do what you want. ... View more

Hi Jeanee, You hit the nail on the head!  When you run Moran's I and do not choose to provide a Spatial Weights Matrix from File, then the tool is essentially creating a SWM on the fly based on the options you filled out for Conceptualization of Spatial Relationships, Distance Method, Standardization, and Distance Band or Threshold Distance.  All of those options enable the tool to define spatial relationships on the fly.  There are some additional options for defining spatial relationships when you use the Generate Spatial Weights Matrix tool, so it is really up to you whether you need to use a SWM or you can define your relationships adequately using the on-the-fly methods. ... View more

Hi A, Those are two great questions!  As far as the F-Statistic for GWR, the answer to this question is actually one of the main reasons that we recommend so strongly that you find a properly specified OLS model before moving on to GWR.  OLS provides tons of great diagnostics that can help you figure out if you've met all of the assumptions of OLS.  Those diagnostics are what help us feel so confident that the model that we've found really is a model that we can trust.  To learn more about those assumptions check out the ArcUser article called Finding a Meaningful Model.  Unlike OLS, however, GWR does not have many of those great diagnostics, meaning it is a lot more difficult to figure out if you've found a model that you can trust.  The F-Statistic, as well as most of the other diagnostics, are not available for a GWR analysis.  It is for this reason that it is so important to find a model that meets all of the criteria using OLS before you move on to GWR. As far as mapping the Adjusted R-Squared values, Adjusted R-Squared is a global value that applies to the entire study area (the entire model).  As a result, there is only one Adjusted R-Squared value for the analysis, meaning that there is nothing to map.  The only value that is output for each individual feature is the local R-Squared value, which explains why you can map the local R-Squared, but not the Adjusted R-Squared. ... View more

Hi Mareike, I'm really sorry that you're having trouble! Would it be possible for you to attach a copy of the graphical output from K-Function?  That would help us start to figure out exactly what's going on.  Hopefully we'll be able to figure this out and get you up and running! Lauren Rosenshein Geoprocessing Product Engineer ... View more

Hi Remi, I'm really sorry that you're having trouble with the supplementary spatial statistics tools!  In order for us to try and reproduce the issue, can you please tell us exactly what operating system, version of ArcGIS, and regional language settings are.  This way we can set up a machine that matches yours and try to reproduce the issue.  It would also be very helpful if you could try changing your regional language settings to English.  If you still are having trouble running the tools (meaning that localization issues are not the problem), then we will have to investigate further.  Either way, thank you very much for bringing this to our attention.  It will help us make sure that these tools work as expected when they are released officially with 10.1, and hopefully we'll be able to help you get them running on your own machine in the meantime. Lauren Rosenshein Geoprocessing Product Engineer ... View more

Hi Juan, I'm glad that you found the tutorial useful.  Actually, there are quite a few ways to dig deeper into the algorithms and methods behind the hot spot analysis tool.  For one thing, the documentation called How Hot Spot Analysis Works outlines the algorithms that are used for a hot spot analysis.  Another great way to learn more about the tools, especially if you are interested in implementing something similar on your own, is to actually look at the source code associated with the tools.  Most of the tools in the Spatial Statistics toolbox are delivered as python scripts.  From the ArcToolbox Window, if you right-click on the Hot Spot Analysis tool and choose "Edit" you can actually see the code, which is a great way to learn more about the method. As far as your questions about the definitions, could you be a little more specific about the context of the question?  Where did you see those terms used?  Knowing where you saw those terms used will help be figure out the best way to explain them in context. ... View more