Detransformation bias in nonlinear trip generation models

Liming Wang, Kristina M. Currans

Research output: Contribution to journalArticle

Abstract

In recent years, there have been substantial efforts from researchers and practitioners to improve site-level trip generation estimation methods to address some of the pitfalls of conventional approaches for applications such as traffic impact analyses. These new trip generation models often adopt sophisticated nonlinear model forms to utilize new information and incorporate new factors influencing trip generation. However, if sufficient caution is not taken in their application, these new predictive models may introduce severe bias. This paper focuses on a typical source of biases in the applications of such models arising from detransformation of predictions from models with a nonlinearly transformed dependent variable in the prediction process (for example, predicting from a semilog model). While such biases are well known and corrections have been proposed in other disciplines, they have not been adopted in site-level trip generation models to the authors' knowledge. The detransformation bias is described and demonstrated-focusing on log-transformed models-with numeric simulations and empirical studies of trip generation models, before discussing their implications for trip generation applications and research.

Original languageEnglish (US)
Article number04018021
JournalJournal of Urban Planning and Development
Volume144
Issue number3
DOIs
StatePublished - Sep 1 2018

Fingerprint

trend
non-linear model
predictive model
estimation method
prediction
traffic
simulation

Keywords

  • Bias
  • Development-level estimation
  • Land-use development
  • Predictive model
  • Traffic impact analyses
  • Transportation impact analyses
  • Trip generation

ASJC Scopus subject areas

  • Civil and Structural Engineering
  • Geography, Planning and Development
  • Development
  • Urban Studies

Cite this

Detransformation bias in nonlinear trip generation models. / Wang, Liming; Currans, Kristina M.

In: Journal of Urban Planning and Development, Vol. 144, No. 3, 04018021, 01.09.2018.

Research output: Contribution to journalArticle

@article{4a747d7cd43b4497b0747c1b9cf703fb,
title = "Detransformation bias in nonlinear trip generation models",
abstract = "In recent years, there have been substantial efforts from researchers and practitioners to improve site-level trip generation estimation methods to address some of the pitfalls of conventional approaches for applications such as traffic impact analyses. These new trip generation models often adopt sophisticated nonlinear model forms to utilize new information and incorporate new factors influencing trip generation. However, if sufficient caution is not taken in their application, these new predictive models may introduce severe bias. This paper focuses on a typical source of biases in the applications of such models arising from detransformation of predictions from models with a nonlinearly transformed dependent variable in the prediction process (for example, predicting from a semilog model). While such biases are well known and corrections have been proposed in other disciplines, they have not been adopted in site-level trip generation models to the authors' knowledge. The detransformation bias is described and demonstrated-focusing on log-transformed models-with numeric simulations and empirical studies of trip generation models, before discussing their implications for trip generation applications and research.",
keywords = "Bias, Development-level estimation, Land-use development, Predictive model, Traffic impact analyses, Transportation impact analyses, Trip generation",
author = "Liming Wang and Currans, {Kristina M.}",
year = "2018",
month = "9",
day = "1",
doi = "10.1061/(ASCE)UP.1943-5444.0000455",
language = "English (US)",
volume = "144",
journal = "Journal of the Urban Planning and Development Division, ASCE",
issn = "0733-9488",
publisher = "American Society of Civil Engineers (ASCE)",
number = "3",

}

TY - JOUR

T1 - Detransformation bias in nonlinear trip generation models

AU - Wang, Liming

AU - Currans, Kristina M.

PY - 2018/9/1

Y1 - 2018/9/1

N2 - In recent years, there have been substantial efforts from researchers and practitioners to improve site-level trip generation estimation methods to address some of the pitfalls of conventional approaches for applications such as traffic impact analyses. These new trip generation models often adopt sophisticated nonlinear model forms to utilize new information and incorporate new factors influencing trip generation. However, if sufficient caution is not taken in their application, these new predictive models may introduce severe bias. This paper focuses on a typical source of biases in the applications of such models arising from detransformation of predictions from models with a nonlinearly transformed dependent variable in the prediction process (for example, predicting from a semilog model). While such biases are well known and corrections have been proposed in other disciplines, they have not been adopted in site-level trip generation models to the authors' knowledge. The detransformation bias is described and demonstrated-focusing on log-transformed models-with numeric simulations and empirical studies of trip generation models, before discussing their implications for trip generation applications and research.

AB - In recent years, there have been substantial efforts from researchers and practitioners to improve site-level trip generation estimation methods to address some of the pitfalls of conventional approaches for applications such as traffic impact analyses. These new trip generation models often adopt sophisticated nonlinear model forms to utilize new information and incorporate new factors influencing trip generation. However, if sufficient caution is not taken in their application, these new predictive models may introduce severe bias. This paper focuses on a typical source of biases in the applications of such models arising from detransformation of predictions from models with a nonlinearly transformed dependent variable in the prediction process (for example, predicting from a semilog model). While such biases are well known and corrections have been proposed in other disciplines, they have not been adopted in site-level trip generation models to the authors' knowledge. The detransformation bias is described and demonstrated-focusing on log-transformed models-with numeric simulations and empirical studies of trip generation models, before discussing their implications for trip generation applications and research.

KW - Bias

KW - Development-level estimation

KW - Land-use development

KW - Predictive model

KW - Traffic impact analyses

KW - Transportation impact analyses

KW - Trip generation

UR - http://www.scopus.com/inward/record.url?scp=85046686214&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85046686214&partnerID=8YFLogxK

U2 - 10.1061/(ASCE)UP.1943-5444.0000455

DO - 10.1061/(ASCE)UP.1943-5444.0000455

M3 - Article

AN - SCOPUS:85046686214

VL - 144

JO - Journal of the Urban Planning and Development Division, ASCE

JF - Journal of the Urban Planning and Development Division, ASCE

SN - 0733-9488

IS - 3

M1 - 04018021

ER -