U.S. flag

An official website of the United States government, Department of Justice.

NCJRS Virtual Library

The Virtual Library houses over 235,000 criminal justice resources, including all known OJP works.
Click here to search the NCJRS Virtual Library

How To Handle Seasonality - Introduction to the Detection and Analysis of Seasonal Fluctuation in Criminal Justice Time Series

NCJ Number
92928
Author(s)
C R Block
Date Published
1983
Length
68 pages
Annotation
This report reviews the component and stochastic approaches to analyzing seasonality of crime and gives the analyst basic guidelines for using them in practical situations.
Abstract

Two kinds of information may be useful in making decisions: a description of seasonal fluctuation and a description of the variation in a series with the seasonal fluctuation removed. The two major approaches to defining seasonability are mathematically similar, but have practical differences. The component method emphasizes a separate description of seasonal fluctuation, while the stochastic approach emphasizes forecasting the future with a model that incorporates seasonality. Thus, the former focuses on seasonality itself, while the latter addresses seasonality as it affects the accuracy of a forecast. Component methods discussed include the moving average, additive/multiplicative assumption, the F of stable seasonality, the relative contribution of the irregular, the average duration of run, months for cyclical dominance, pattern consistency, and trading day option. Stochastic methods explained are moving average and autoregressive processes, identifying the process of a series, stationarity, and model evaluation. Appropriate applications of each method are described. The report notes that choice of a method depends on the study's objectives and the analyst's judgment. Tables and graphs are included. An annotated bibliography of 110 references is appended.