This initial clustering model analysis is exploring agricultural commodity loss
nationwide, and how damage causes may be similar or dissimilar per commodity and year.  Our steps:

  1. We aggregate and summarize all insurance loss for an individual commodity for a set of years, creating a sparse matrix of damage causes for each county. We initially estimate the optimal number of clusters using NBClust to pick the optimum number of clusters by using a range of methods and then selecting bed option.
  2. Using this optimal cluster number, we visually examine our clusters (again, which are for a singular commodity, per year, at a county level) using a PCA based 2 dimensional plot (using PC1 and PC2).  We then map our clusters spatially.  The cluster plot lists the amount of variance for each PC.
  3. The example plots and maps below are for corn and wheat, for water scarcity damage causes (drought, heat, excessive sun, fire, hot wind) and cold (cold wet weather, cold winter, freeze, frost).  Again, this methodology can be run for any commodity across the full time frame range of insurance claims.  Finally, the below analysis is for loss per acre ($).

 

RMarkdown code that generates the following analyses:

agloss unsupervised lossperacre water scarcity.Rmd

agloss unsupervised lossperacre cold.Rmd

Analysis Results

WHEAT 2009 – Water Scarcity

WHEAT 2011 – Water Scarcity

WHEAT 2015 – Water Scarcity

CORN 2009 – Water Scarcity

CORN 2011 – Water Scarcity

CORN 2015 – Water Scarcity

CORN 2009 – Cold

CORN 2011 – Cold

Summary Results Plots

Unsupervised paper3 example plots