The first thing I did was pull some data from the United States Bureau of Economic Analysis and hustled a quick heatmap to get an idea of what the data looks like. By doing so we can see that there are annual trends for unemployment, and it becomes more apparent where the periods of recessions are grouped.
The next step was to put this in context and try to uncover trends. This involved fitting the data to a time series and attaching a LOESS curve along with periods of recession extrapolated from the heatmap. This does a good job of showing the spikes of unemployment relative to recessions, and the overall increasing rate of unemployment since 1948. It also reveals to us that in terms of unemployment the recession of the early 1980's was worse than the most recent recession.
The last step was to break this information down to the county level and track the spatial movement and shape of unemployment. The time frame to do this was condensed a bit to 2008-2011, but the effects are interesting to say the least. The data shows that unemployment woes began in the urban areas, then spread into the interior. Conversely, it also appears that the recovery began in urban area as well, and has been diffusing inland from there.
For my next section I'm going to break this information down a bit more and provide a deeper analysis.