Monitoring Temporal Crime Trends

The main reason crime analysts monitor crime trends over time is to identify anomalous spikes. For example, if there are more burglaries this month than you expected, you may look more closely into those events to identify if they are connected to a single burglar going on a spree of events. This webinar discusses metrics to identify when crime is higher than you would expect and provides examples of creating simple rules (e.g. more than 5 events in a week), identifying if an increase is within the usual range in CompStat reports, and creating graphs to show short term spikes and long term trends.

Dr. Andrew Wheeler has a PhD in criminal justice from SUNY Albany (2015), and has worked with police departments and different criminal justice agencies across the United States. During his graduate studies, he worked as a crime analyst at Troy, NY police department. Afterward, he was a professor of criminology and is currently employed as a data scientist in the private sector. His work can be seen at https://crimede-coder.com/