Research Mentor(s)
Noguchi, Kimihiro
Description
Our purpose is to use time series analysis to model and forecast the underlying dynamics behind crime in Bellingham, Washington. Using recent monthly data from the Bellingham Police Department, we considered singular spectrum analysis and autoregressive moving average modelling techniques to estimate significant deterministic patterns in the data. After examining the multitude of data provided, we narrowed down to two categories of crime: alcohol offenses and domestic violence. We created two time series models for each category and compared them to each other. The better performing model was used to forecast the number of crime incidents for ten months and identify trends and seasonality.
Document Type
Event
Start Date
16-5-2018 12:00 PM
End Date
16-5-2018 3:00 PM
Department
Mathematics
Genre/Form
student projects, posters
Subjects – Topical (LCSH)
Time-series analysis; Criminal statistics--Washington (State)--Bellingham
Geographic Coverage
Bellingham (Wash.)
Type
Image
Keywords
Crime, crime research, time series, forecasting, domestic, violence, alcohol, ARIMA, SSA, Bellingham, trend, seasonality
Rights
Copying of this document in whole or in part is allowable only for scholarly purposes. It is understood, however, that any copying or publication of this document for commercial purposes, or for financial gain, shall not be allowed without the author’s written permission.
Language
English
Format
application/pdf
Included in
Modeling and Forecasting Crime Patterns in Bellingham, Washington
Our purpose is to use time series analysis to model and forecast the underlying dynamics behind crime in Bellingham, Washington. Using recent monthly data from the Bellingham Police Department, we considered singular spectrum analysis and autoregressive moving average modelling techniques to estimate significant deterministic patterns in the data. After examining the multitude of data provided, we narrowed down to two categories of crime: alcohol offenses and domestic violence. We created two time series models for each category and compared them to each other. The better performing model was used to forecast the number of crime incidents for ten months and identify trends and seasonality.