Latest Contributions by LaurenRosenshein

Hi Alice, That's a great question.  We do have some great tools for doing regression analysis inside of ArcGIS, although they were not implemented until ArcGIS 9.3. Whether you have access to a later version of ArcGIS or not, there are tons of resources to help you get started with regression analysis, which you can find here: http://esriurl.com/spatialstats.  You may want to start with the Training Seminar called Regression Analysis Basics.  There is also an article that we wrote recently called Finding a Meaningful Model which talks about some of the ins and outs of finding a good model.    You will need to sample your data (using any variety of methods, including the Sample tool) to perform your regression analysis (whether you perform your analysis in ArcGIS or another package), which is important with raster data because of the spatial autocorrelation between the cells. There are tons of different software packages that you can use to do this type of regression analysis, so even if you are unable to get access to ArcGIS 9.3 or later and need to do your regression analysis in another package, I hope that these resources help you with your analysis! ... View more

Hey Bethany and elsetton, This is a really good question!!  The answer is actually pretty simple.  The overall R2 value for GWR is calculated the exact same way that it is calculated in OLS...essentially it's the proportion of the variability explained by the model.  For both it is 1 - (Σ(predicted-observed)²/Σ(observed-mean of all observed)²).  In other words, its 1 minus the variance of the residuals divided by the variance of the input data.  In both OLS and GWR, the residuals are the predicted minus the observed, so it follows that you would expect OLS and GWR to have different R-squared values because those predicted values are calculated using different methods.  So, its the same as OLS, the only difference is what values are being used for the predictions. This is probably way more than you wanted to know, but hopefully it helps!  Let me know if you need any more details.  🙂 Lauren Rosenshein Geoprocessing Product Engineer ... View more

Hey Andrew, I'm sorry you're having trouble getting GWR to run for your analysis.  Severe model design errors often indicate a problem with global or local multicollinearity. To determine where the problem is, run the model using OLS and examine the VIF value for each explanatory variable. If some of the VIF values are large (above 7.5, for example), global multicollinearity is preventing GWR from solving. More likely, however, local multicollinearity is the problem. Try creating a thematic map for each explanatory variable. If the map reveals spatial clustering of identical values, consider combining those variables with other explanatory variables to increase value variation. If, for example, you are modeling home values and have variables for both bedrooms and bathrooms, you may want to combine these to increase value variation or represent them as bathroom/bedroom square footage. Another option is to try transforming it (although not in the traditional sense of logs or powers): create a new field, then calculate the values to be the value (in this case the log) minus the mean for all values in that field.  This doesn�??t actually change anything (the impact on results), but for some reason we've found that GWR likes variables in that form�?� and this transformation will often �??fix�?� the problem. Also, just a reminder to make sure that you find a properly specified OLS model before moving on to GWR.  There is some great documentation about this, including this recent ArcUser article on Finding a Meaningful Model and the training seminar on Regression Analysis Basics. ... View more

Hi A, I'm so sorry that you've run into trouble with your system, and subsequently with ArcToolbox.  Unfortunately, the best advice that I can think of is to reinstall ArcMap, although I totally understand your worries about being able to reinstall with your student license.  If the only tools you're having trouble with are python scripts, you may want to try reinstalling Python itself...but that will only be helpful if the problem is isolated to python scripts (and even then, its just a shot in the dark).  I think you're best bet may be to try and call tech support to see if they have any ideas about what you can do.  Sorry I couldn't be more helpful! Lauren Rosenshein Geoprocessing Product Engineer ... View more

Hi George, I'm very sorry that you're having trouble running through the tutorial!  We've runt through those tutorials quite a few times, and have not been able to reproduce your error, so what we have to do is try to figure out what's different about your machine/system that is causing problems.  Can you please let us know about any regional settings that might be different or any other unique system settings that we might be able to set on our own machines to try and reproduce the problem.  That would be very helpful for us.  We absolutely want to help you be successful running through the tutorials, so we will do our very best to get this figured out! Once again, I'm very sorry you're having trouble! (But I'm very glad that you've found our resources useful!)  🙂 Lauren Rosenshein Geoprocessing Product Engineer ... View more

Hi Pei-Fen, The first thing I want to make sure I understand is what kind of field you're using for the Hot Spot Analysis using the Getis-Ord Gi* statistics.  The Hot Spot Analysis tool requires that you have an analysis field, so if you don't have one, you'll have to start by aggregating your incident data.  There are a number of ways to do that (including the grid option that you mentioned, which may be a good one), all discussed here.  Once you've aggregated your data, then you do have to move towards defining your spatial relationships.  One thing that you might want to consider, since it looks like your data is all along a road network, is creating a Network Spatial Weights Matrix, and using the K Nearest Neighbors conceptualization, and choosing 8 neighbors.  That option would 1) take into consideration the network relationships of your features, and 2)ensure every feature had 8 neighbors.  Even if you don't want to use a network, creating a normal Spatial Weights Matrix, and using the k nearest neighbors option, might be good.  As far as the difference between a kernel density analysis and the Getis-Ord Gi* statistics, they are really answering two different questions.  A kernel density analysis will give you an idea of how many features are in a given area, making it easy to find places with lots of features and areas with fewer features.  The hot spot analysis tool is looking at the value of each feature and its neighboring features, comparing that to the values for global dataset, and figuring out if the difference between the two is significant.  When the values are significantly higher for a neighborhood than for the global dataset, then its a hotspot.  If you have one very high value surrounded by much lower values, then it will depend greatly on the global average whether or not that is considered a hot spot or not (If you're interested in finding outliers, you may want to consider the Cluster and Outlier Analysis tool).  When the values are significantly lower for a neighborhood than for the global dataset, then its a cold spot.  Then, what's red on the map and what's blue on the map is based on statistical significance, as opposed to a density surface which can be relatively subjective.  You may want to check out this video: Performing Proper Density Analysis. ... View more

Hi balbright76, That's a great question!  Unfortunately, there is no way right now to do a hot spot analysis in 3D, although its a great suggestion and definitely something that we have on our list for consideration. Thanks! Lauren Rosenshein Geoprocessing Product Engineer ... View more

Hi Ken, I'm very sorry you're having trouble running OLS.  Would it be possible to provide the shapefile you're using, rather than the excel spreadsheet, so that we can try to reproduce the problem? Thanks in advance, and again, sorry you're having trouble! Lauren Rosenshein Geoprocessing Product Engineer ... View more

Hi!  It looks like you are spelling "EUCLIDEAN" wrong in your code, with an I at the end instead of an E. ... View more

Hi zonabb, The Average Nearest Neighbor tool in 10.0 does create the html report, but the option to create the report is an optional parameter that you have to check on.  The option is called "Generate Report", and if it isn't checked on the report will not be generated. Hope this helps. Lauren Rosenshein Geoprocessing Product Engineer ... View more

Hi Phil, I answered some of this question here: http://forums.arcgis.com/threads/31142-Differences-between-Local-Spatial-Statistics-Results.  The other part of your question is if these different analyzes should be used separately.  My answer would be that they are getting at slightly different questions, so it may be useful to use them both.  There is no reason that they cannot both be used in the same analysis since they both provide complementary insight into the questions that you're asking.  Alternatively, you can just use the one that proves your point or provides evidence for your hypothesis...but trying them both can't hurt. Hope this helps. Lauren Rosenshein Geoprocessing Product Engineer ... View more

Hi Bilal, You are right, you do need at least 80 observations to use GWR (the more the better).  Unfortunately, having 30 observations at each of the 5 stations does not do the trick.  You really need to have more stations.  And the geographic extent of the GWR analysis is always going to be the same as the geographic extent of the input features.  GWR is not an interpolation method.  If you want to do interpolation, you may want to think about something like Kriging, in either Spatial Analyst or Geostatistical Analyst. You can learn more about regression analysis using OLS and GWR, and other spatial statistics tools, here: http://esriurl.com/spatialstats. Hope this helps. Lauren Rosenshein Geoprocessing Product Engineer ... View more

Hi Zahi, This is a great question!  There is absolutely no reason that you can't use GWR to analyze road segments.  As long as each of the centroids that you're analyzing have all of the necessary dependent and explanatory variables, you can use OLS and GWR to analyze them.  With all of the tools in the Spatial Statistics toolbox, points, lines and polygons can be used as the unit of analysis.  For lines and polygons, most of the time the centroid of the feature is being used for analysis (for instance if a fixed distance band is used to determine spatial relationships, that distance is from the centroid of a line or polygon, not the edge).  The major exception is when polygon contiguity is used.  That being said, its important to remember that you always want to start out your regression analysis by finding a properly specified model using OLS.  Once you find a properly specified model using OLS you can feel confident in the results that you find using GWR.  What does it mean to be properly specified?  Well, there are several assumptions of OLS that need to be met, and they are all explained in this ArcUser article called Finding a Meaningful Model.  There is also a Regression Analysis Tutorial, and a free training seminar called Regression Analysis Basics.  There are a bunch of other resources about the spatial statistics also available here:  http://esriurl.com/spatialstats.  Hope this helps. Lauren Rosenshein Geoprocessing Product Engineer ... View more

Hi Mark, This sounds like a very interesting analysis!  One possible way to think about this would be to do a hot spot analysis of the drive times to the nearest site of service.  This would give you an idea of where you have hot spots of customers with very high drive-times to sites of service.  A good place to start.  To learn more about hot spot analysis inside of ArcGIS you may want to check out the short videos and tutorials available here:  http://esriurl.com/spatialstats.  As far as ranking the hot spots, one possibility may be to rank them by the number of features involved?  A hot spot of 10 customers with long drive-times to sites may be less important than a hot spot with 1000 customers with long drive-times. You may also want to think about posting this question in the Network Analyst forum because, as with any complex problem/question, there are always many ways to go about it, and probably the best way to do it is a combination of several methods including some network analysis/optimization analysis. Hope this helps. Lauren Rosenshein Geoprocessing Product Engineer ... View more

Hi Kami, Examining the relationships between the variables that you have in your data can be done using Ordinary Least Squares regression and Geographically Weighted Regression, both available in ArcGIS.  To get started using OLS and GWR, you may want to check out the Regression Analysis Basics seminar, which is free from the Virtual Campus website. You may also want to check out the other resources available for spatial statistics here:  http://esriurl.com/spatialstats. Hope this helps. Lauren Rosenshein Geoprocessing Product Engineer ... View more