GIS5935 Module 12-- Geographically Weighted Regression

-
This week we were introduced to Geographically Weighted Regression (GWR). The main differences between a GWR and an Ordinary Least Squares (OLS) regression are:

  • Where OLS is a global model, GWR is local
  • Distances are NOT considered in OLS, whereas distances define the data fitting GWR
  • Coefficients are constant in OLS, while in GWR they vary by location
I was a lot more comfortable making these regressions and comparing for best fit variable models this time around. The lab consisted first of an example in comparing an OLS and GWR analysis and learning about the significance of differences in their prediction values.


The second part another OLS and GWR analysis, this time I worked somewhat 'off-rails'. The exercise started by selecting a crime type, I chose 'Residential burglary', then making it a feature class in order to continue an analysis to find out what types of variables had the strongest correlation to high 'Residential burglary' rates in this area. The following variables resulted in high correlation values: Population, Age 5-17, Age 40-49, Age 50-64, Median Income. Results from my OLS regression analysis determined that the variable Age 40-49 had the strongest correlation to high rates in 'Residential burglary' crimes.






Comments

Post a Comment

Popular posts from this blog

GIS5100 Week10: Final Project

-
Image
My final project was dedicated to finding suitable areas for a new Elementary School site in Riverside County, California. I started out by compiling 5 pieces of key datasets: County Assessor parcel data, California Schools, Fault zones, Roads, and Public sites. First I performed euclidean distance analyses on four of them in order to produce zonal raster for preferred areas. A weighted analysis based on prioritizing safety shows that ideal areas fall within Hemet, Menifee, and Palm Springs Unified School Districts. Final analysis results show how active school sites are distributed, where all vacant parcels of at least 5 Acres or more lie, and recommended areas for a new school site. The two circled were determined to be the most suitable after analyzing results from weighted overlay analyses and locations of parcels. Additional setbacks are anticipated as more conditions are implemented to the list of mandatory criteria needed to finalize parcels. Looking b
Read more

Module 5- Analytics

-
Image
Now that we've developed a basic comprehension of color themes, labeling, and typography. This week we dived into creating an infographic using data based on the County Health Rankings data from 2018. This is my very first attempt at an infographic, I don’t think it turned out very well. But I learned a lot overall on how to think more artistically, which is not common. I thought about color themes, placement of data, and amount of information to include and in what format (chart, text, map?). It gave me a lot of ideas on how to tackle the next infographic. Another thing I considered was visual balance by seeing how things looked like when aligned with certain ways and combinations. Given more time and tools on this assignment, I would take advantage of software such as Photoshop to include a fun background that compliments the information supplied via ArcGIS Pro.
Read more

Module 3- Terrain Visualization

-
Image
In this week's module, we covered various terrain visualization techniques, including contour lines, hillshade manipulation, and the use of digital elevation models (DEM) on a 2-dimensional surface. The main exercise required to overlay a land cover layer of Yellowstone over an elevation raster layer. Seen below is my final result. On the left is land cover, with some transparency, over a traditional hillshade effect. I selected a traditional effect because it best represented land cover features. On the right is the elevation with a multispectral hillshade effect.  I went through a few color ramps to test which best complemented this DEM's physical features. I also went ahead and grouped many of the original land cover features into Pine Trees, Douglas Firs, etc. in order to generalize the different types of tree types. As for the map layout, at first I only had a single map frame and tested ways to represent elevation alongside land cover since I would otherwise have to mak
Read more