00728420

Component

https://doi.org/10.3141/1542-01

http://scholar.google.com...wart&publication_year=1996

In recent years a number of nonparametric regression-type statistical procedures have been developed. Classification and regression trees (CART) is one such method that can be used as a classifier for a discrete-valued response variable. Advantages of CART over many other methods are its ability to include a relatively large number of independent variables and to identify complex interactions among these variables. A brief description of the CART procedure is given, and its application as a classifier and as a regression model to highway safety analyses is illustrated.

This paper appears in Transportation Research Record No. 1542, Statistical Methods and Accident Analysis for Highway and Traffic Safety.

English

p. 1-5

1996

Transportation Research Record

0309059135

Figures (1) ; References (4) ; Tables (3)

Crash data; Crashes; Highway safety; Regression analysis; Statistics; Variables

Classification and regression trees; Interactions

Data and Information Technology; Highways; Safety and Human Factors; I81: Accident Statistics

TRIS, TRB

Nov 13 1996 12:00AM

• BAYESIAN ASSIGNMENT OF COVERAGE COUNT LOCATIONS TO FACTOR GROUPS AND ESTIMATION OF MEAN DAILY TRAFFIC
• DETECTION OF SAFETY DETERIORATION IN A SERIES OF ACCIDENT COUNTS
• IDENTIFICATION OF SITES WITH PROMISE
• METHODOLOGY FOR IDENTIFYING HIGHWAY SAFETY PROBLEM AREAS
• PITFALLS OF USING R-SQUARED TO EVALUATE GOODNESS OF FIT OF ACCIDENT PREDICTION MODELS
• RATE-QUALITY CONTROL METHOD OF IDENTIFYING HAZARDOUS ROAD LOCATIONS
• RELIABILITY OF STATISTICAL ROAD ACCIDENT INJURY SEVERITY MODELS
• STATISTICAL TEST OF DIFFERENCE BETWEEN EXPECTED ACCIDENT FREQUENCIES