Spatial Statistics Resources

The Spatial Statistics toolbox contains statistical tools for analyzing spatial distributions, patterns, processes, and relationships.  While there may be similarities between spatial and non-spatial (traditional) statistics in terms of concepts and objectives, spatial statistics are unique in that they were developed specifically for use with geographic data.  Unlike traditional non-spatial statistical methods, they incorporate space (proximity, area, connectivity, and/or other spatial relationships) directly into their mathematics.

There are a ton of resources about using and understanding Spatial Statistics, and here on the Spatial Statistics team we want to make sure that everyone knows where to find them. We also want to make sure that everyone knows that if you have ArcGIS, you already have the Spatial Statistics toolbox, as well as the source code for most of the tools in it.  The Spatial Statistics tools are not an extension, and with the exception of only 2 tools*, every tool in the toolbox is available at all license levels!

We’ve created a shortcut link to the blog that is easy to remember: http://esriurl.com/spatialstats.  If you have any questions, there is now a Spatial Statistics Forum on the Resource Center.  It’s a great place to post your spatial stats questions so that everyone can benefit from the responses!

If you’re looking for the Space-Time Cube Utilities toolbox for use with ArcGIS Pro, we will make it available at the same time that 10.3 releases. Stay tuned!

Last update: December 10th, 2014

 

Technical Workshops from the 2014 International User Conference

1. Spatial Statistics: Simple Ways to Do More with Your Data (VideoPDF)

2. Spatial Data Mining: A Deep Dive into Cluster Analysis (VideoPDF)

3. Beyond Where: Modeling Spatial Relationships Using Regression Analysis (VideoPDF)

4. Applying Spatial Statistics: The Analysis Process in Action (VideoPDF)

 

Spatial Pattern Analysis: Mapping Trends and Clusters

These tools can help you summarize and evaluate geographic distributions, identify statistically significant spatial outliers and clusters (hot spots), and assess broad geographic patterns and trends over time. These resources will help you find patterns and relationships in your data, facilitating discussion, contributing to research, and informing decision making.

Short Videos:

Tutorials/Training

Model and Script Tools

Presentations

Online Documentation

Articles and Blogs

Modeling Spatial Relationships Using Regression Analysis

Regression analysis helps you examine, model, and explore data relationships.  Ultimately, regression analysis helps you answer “why?” questions: “Why do we see so much disease in particular areas?”, “What are the factors that contribute to consistently high childhood obesity rates?”, and “Why are screening rates so low in particular regions of the country?”. Regression analysis also allows you to predict spatial outcomes for other places or other time periods: “How will improvements to road conditions impact traffic fatalities?” or “How will projected population growth affect the demand for health services?”.  These resources will help you learn about basic regression analysis concepts and workflows as they relate to the analysis of geographic data.  Learn how to build a properly specified OLS model, interpret regression results and diagnostics, and potentially use the results of regression analysis to design targeted interventions.

Tutorials/Training

Model and Script Tools

Presentations

Online Documentation

Articles and Blogs

General Resources for Spatial Statistics Users

Presentations

Model and Script Tools

Articles and Blogs

Books

  • Fotheringham, Stewart A., Chris Brunsdon, and Martin Charlton. Geographically Weighted Regression: the analysis of spatially varying relationships. John Wiley & Sons, 2002.

*Geographically Weighted Regression (GWR) requires either an ArcInfo License OR the Spatial Analyst Extension OR the Geostatistical Analyst Extension
*Generate Network Spatial Weights requires the Network Analyst Extension

This entry was posted in Analysis & Geoprocessing and tagged , , , . Bookmark the permalink.

Leave a Reply

5 Comments

  1. eulessdave says:

    Also, there are two books from ESRI Press on using the Spatial Analysis and Spatial Statistics tool. It’s the same book, but the first is the ArcGIS 9.3 version and the second is the ArcGIS 10 version.

    GIS Tutorial II (9.3) -
    http://esripress.esri.com/display/index.cfm?fuseaction=display&websiteID=153&moduleID=0
    GIS Tutorial 2 (10)-
    http://esripress.esri.com/display/index.cfm?fuseaction=display&websiteID=185&moduleID=0

  2. sperrys says:

    The Geographic Weighted Tool in 10.1 cannot write out a raster coefficient files. Everything else works. Error is the infamous “ERROR 999998: Unexpected Error”. It can’t create output raster. It completes the first coefficient file, but the program crashes about 20% into the second coefficient file and deletes all files. I was able to examine the first file once before it crashed. It did not have any useful data or use use the environment settings. Has anyone made it work?

  3. mtuffly says:

    Dear ESRI I have a question regarding the creation of a spatial weight matrix (SWM). I have a data set that contains 71 points depicting ozone values for two time periods (n = 142). When I create a SWM for each time period separately (i.e. independent of time) and run Moran’s I I get the same results as my R Morans’ I. That is ArcGIS and R produce the same results in Morans’ I; hence both methods must porduce the same SWM. So this is a good check.

    Now here is where things get interesting. When I create a SWM in ArcGIS using the concept of TIME_SPACE and run Morans’I I get different results when compared to my R program using Moran’s I. Since I have concluded that Moran’s I is calculated the same in both my R program and ArcGIS the issue must lies in the generation of the SWM under the concept TIME_SPACE. So my questions is how does ArcGIS combine the matrices generated from the 71 points over two time periods. My SWM generated in R is 142 rows by 142 columns. If I take the ArcGIS SWM convert it to a tables I get 11534 records. If my guess is true (e.g. every table record is a element in the matrix) then ArcGIS does not create a symmetrical matrix. That is the root of 11534 is 107.4

    In simple terms how are the intput matrix for the two time periods combine to create a single SWM.

    Thanks
    Mike

  4. janikas says:

    It should be symmetric… but… the sum of the non-zero elements of the weights matrix does not have to have an even square in order to be symmetric. It should be even… which it is… but symmetry would need to be verified separately using linear algebra on a full matrix or by using read/neighbor check in Python. I could help you further offline if you want and we can get to the bottom of it… you could elaborate on which functions you are using in R and I could give you some Python code to test the symmetry of a SWM file. You can reach me at mjanikas@esri.com.

  5. mtuffly says:

    Question about Geographic Weighted Regression (GWR)
    If I have 73 points (n = 73) and I run GWR using these 73 points and the CV method coupled with my covariates (3 in number). I get 73 unique equations with four coefficients. When I use the option to create the surface of the GWR coefficients the GWR tool creates the correct number of surfaces associated with the number of covariates and intercept.

    To be specific what is the decay function used in the output. That is, how are the cell values calculated (decayed) between my observed points.

    Thanks

    Mike