Transport Data Analysis and Modeling Methodologies

Similar documents
Queensland University of Technology Transport Data Analysis and Modeling Methodologies

Phd Program in Transportation. Transport Demand Modeling. Session 11

Discrete Choice Modeling

Multi-Vehicle Crashes Involving Large Trucks: A Random Parameter Discrete Outcome Modeling Approach

WesVar uses repeated replication variance estimation methods exclusively and as a result does not offer the Taylor Series Linearization approach.

Valuing Environmental Impacts: Practical Guidelines for the Use of Value Transfer in Policy and Project Appraisal

Didacticiel - Études de cas. In this tutorial, we show how to implement a multinomial logistic regression with TANAGRA.

Econometric Methods for Valuation Analysis

Actuarial Research on the Effectiveness of Collision Avoidance Systems FCW & LDW. A translation from Hebrew to English of a research paper prepared by

What is the Level of Volatility in Instantaneous Driving Decisions?

Discrete Choice Modeling William Greene Stern School of Business, New York University. Lab Session 4

GOALS. Discrete Probability Distributions. A Distribution. What is a Probability Distribution? Probability for Dice Toss. A Probability Distribution

Discrete Probability Distributions Chapter 6 Dr. Richard Jerz

Categorical Outcomes. Statistical Modelling in Stata: Categorical Outcomes. R by C Table: Example. Nominal Outcomes. Mark Lunt.

Top Incorrect Problems

Econometrics II Multinomial Choice Models

A Gender-based Analysis of Work Trip Mode Choice of Suburban Montreal Commuters Using Stated Preference Data

Recovery measures of underfunded pension funds: contribution increase, no indexation, or pension cut? Leo de Haan

Unit 04 Review. Probability Rules

Statistical Analysis of Traffic Injury Severity: The Case Study of Addis Ababa, Ethiopia

Probability Distributions. Chapter 6

What is spatial transferability?

Table 4. Probit model of union membership. Probit coefficients are presented below. Data from March 2008 Current Population Survey.

Americans AV Preferences: Dynamic Ride-Sharing, Privacy & Long-Distance Mode Choices. Dr. Kara Kockelman & Krishna Murthy Gurumurthy

WORKING PAPER ITLS-WP Does the choice model method and/or the data matter? INSTITUTE of TRANSPORT and LOGISTICS STUDIES

BEcon Program, Faculty of Economics, Chulalongkorn University Page 1/7

Contents. Part I Getting started 1. xxii xxix. List of tables Preface

Exercise 1. Data from the Journal of Applied Econometrics Archive. This is an unbalanced panel.n = 27326, Group sizes range from 1 to 7, 7293 groups.

Online Appendix for Does mobile money affect saving behavior? Evidence from a developing country Journal of African Economies

sociology SO5032 Quantitative Research Methods Brendan Halpin, Sociology, University of Limerick Spring 2018 SO5032 Quantitative Research Methods

15. Multinomial Outcomes A. Colin Cameron Pravin K. Trivedi Copyright 2006

Appendix. Table A.1 (Part A) The Author(s) 2015 G. Chakrabarti and C. Sen, Green Investing, SpringerBriefs in Finance, DOI /

STATISTICAL METHODS FOR CATEGORICAL DATA ANALYSIS

Estimating Mixed Logit Models with Large Choice Sets. Roger H. von Haefen, NC State & NBER Adam Domanski, NOAA July 2013

Computer Lab II Biogeme & Binary Logit Model Estimation

Available online at ScienceDirect. Transportation Research Procedia 1 (2014 ) 24 35

ARIMA ANALYSIS WITH INTERVENTIONS / OUTLIERS

VALUATION OF TRAVEL TIME SAVING WITH REVEALED PREFERENCE DATA IN JAPAN: FURTHER ANALYSIS

Web Appendix. Are the effects of monetary policy shocks big or small? Olivier Coibion

Module 9: Single-level and Multilevel Models for Ordinal Responses. Stata Practical 1

Egyptian Married Women Don t desire to Work or Simply Can t? A Duration Analysis. Rana Hendy. March 15th, 2010

Drawbacks of MNL. MNL may not work well in either of the following cases due to its IIA property:

Tutorial: Discrete choice analysis Masaryk University, Brno November 6, 2015

Stability of parameters estimated on. Cross-sectional data

A Comparison of Univariate Probit and Logit. Models Using Simulation

Using Halton Sequences. in Random Parameters Logit Models

DYNAMICS OF URBAN INFORMAL

To be two or not be two, that is a LOGISTIC question

Sean Howard Econometrics Final Project Paper. An Analysis of the Determinants and Factors of Physical Education Attendance in the Fourth Quarter

Household Flood Evacuation Route Choice Models at Sub-district Level

Sociology Exam 3 Answer Key - DRAFT May 8, 2007

An Activity-Based Microsimulation Model of Travel Demand in the Jakarta Metropolitan Area

Discrete Probability Distributions

Attachment JEM 4 Hearing Exhibit 116 Page 1 of 11] Residential Sales 5,817,938 90,842,431 96,660,368. Normal HDD

Transportation Research Forum

Introduction to the Maximum Likelihood Estimation Technique. September 24, 2015

Review questions for Multinomial Logit/Probit, Tobit, Heckit, Quantile Regressions

Revealing Additional Dimensions of Preference Heterogeneity in a Latent Class Mixed Multinomial Logit Model

Volvo City Safety loss experience by vehicle age

Southern California Association of Governments (SCAG) Metropolitan Planning Organization (AMPO) Annual Conference. Prepared for

A MODIFIED MULTINOMIAL LOGIT MODEL OF ROUTE CHOICE FOR DRIVERS USING THE TRANSPORTATION INFORMATION SYSTEM

against city bus services. Figure 1. A snapshot of low-service-quality buses with open-end carriages (City bus route 46) To cope with increased traffi

Omitted Variables Bias in Regime-Switching Models with Slope-Constrained Estimators: Evidence from Monte Carlo Simulations

Automobile Ownership Model

Modeling of Shopping Participation and Duration of Workers in Calicut

AIC = Log likelihood = BIC =

FIT OR HIT IN CHOICE MODELS

An Analysis of the Factors Affecting Preferences for Rental Houses in Istanbul Using Mixed Logit Model: A Comparison of European and Asian Side

Lecture 21: Logit Models for Multinomial Responses Continued

POLYTECHNIC OF NAMIBIA SCHOOL OF MANAGEMENT SCIENCES DEPARTMENT OF ACCOUNTING, ECONOMICS AND FINANCE ECONOMETRICS. Mr.

Grouped Data Probability Model for Shrimp Consumption in the Southern United States

WesVar Analysis Example Replication C7

Joint Mixed Logit Models of Stated and Revealed Preferences for Alternative-fuel Vehicles

Mode-choice behaviour for home-based work trips

Probability Distributions. Chapter 6

*9-BES2_Logistic Regression - Social Economics & Public Policies Marcelo Neri

Estimation and Confidence Intervals

Evaluation of influential factors in the choice of micro-generation solar devices

Nonlinear Econometric Analysis (ECO 722) Answers to Homework 4

THE CITY UNIVERSITY OF NEW YORK VEHICLE USE POLICY

Choice Probabilities. Logit Choice Probabilities Derivation. Choice Probabilities. Basic Econometrics in Transportation.

Intro to GLM Day 2: GLM and Maximum Likelihood

Ministry of Health, Labour and Welfare Statistics and Information Department

INSTITUTE OF ACTUARIES OF INDIA EXAMINATIONS. 20 th May Subject CT3 Probability & Mathematical Statistics

Estimation and Confidence Intervals

Supporting Information for:

Hot Springs Bypass Extension TIGER 2017 Application. Benefit-Cost Analysis Methodology Summary

Calibration of Nested-Logit Mode-Choice Models for Florida

Brief Sketch of Solutions: Tutorial 2. 2) graphs. 3) unit root tests

Virginia Department of Education

Project Summary Project Name: Route 37 Corridor Safety Sweep Project Number:

1. 1. Do you usually drive above the 55 mph speed limit posted on urban interstate highways if so by how much?

COMPANY TARIFF FOR THE MOTOR VEHICLE LIABILITY INSURANCE Tariff Provisions (ITALY)

TOURISM GENERATION ANALYSIS BASED ON A SCOBIT MODEL * Lingling, WU **, Junyi ZHANG ***, and Akimasa FUJIWARA ****

Virginia Railway Express Annual Customer Survey Customer Opinion Survey Results

HOW MUCH DOES THAT TRAFFIC TICKET REALLY COST

Unpacking Preference: How Previous Experience Affects Auto Ownership

Emilia Istrate, Senior Research Analyst. July 28, 2009 Washington DC

Discussion on Policy-Relevant Exchange Rate Pass-Through to U.S. Import Prices

Econ 3790: Business and Economics Statistics. Instructor: Yogesh Uppal

Transcription:

Transport Data Analysis and Modeling Methodologies Lab Session #14 (Discrete Data Latent Class Logit Analysis based on Example 13.1) In Example 13.1, you were given 151 observations of a travel survey collected in State College Pennsylvania (See Example 13.1 on page 319 of the text for an estimation of a fixedparameters logit model of these data). All of the households in the sample are making the morning commute to work. They are all departing from the same origin (a large residential complex in the suburbs) and going to work in the Central Business District. They have the choice of three alternate routes; 1) a four-lane arterial (speed limit = 35mph, 2 lanes each direction), 2) a two-lane rural road (speed limit = 35mph, 1 lane each direction) and 3) a limited access four-lane freeway (speed limit = 55mph, 2 lanes each direction). Your task is to experiment with a random parameters and latent class logit model using these data. Your write-up should include: 1. The results of your best model specification. 2. A discussion of the findings in searching for a random parameters specification. Again, for reference, see Example 13.1 on page 319 of the text.

Variables available for your specification are (in file Ex13-1.txt): Variable Number x1 x2 x3 x4 x5 x6 x7 x8 x9 Explanation Route chosen, rows: 1 - arterial, 2 - rural road, 3 - freeway Arterial row indicator; 1 for arterial row, 0 for others Rural row indicator; 1 for rural row, 0 for others Freeway row indicator; 1 for freeway row, 0 for others Traffic flow rate Number of traffic signals Distance in tenths of miles Seat belts: 1 - if wear, 0 - if not Number of passengers in car x10 Driver age in years: 1-18 to 23, 2-24 to 29, 3-30 to 39, 4-40 to 49, 5-50 and above x11 x12 x13 x14 Gender: 1 - male, 0 - female Marital status: 1 - single, 0 - married Number of children Annual income: 1 - less than 20000, 2-20000 to 29999, 3-30000 to 39999, 4-40000 to 49999, 5 - more than 50000 x15 Model year of car (e.g. 86 = 1986) x16 x17 Origin of car: 1 - domestic, 0 - foreign Fuel efficiency in miles per gallon

Random Parameters: --> RESET Initializing NLOGIT Version 4.0.1 (January 1, 2007). --> read;nvar=17;nobs=453;file=d:\old_drive_d\book\book2e-data\ex13-1.txt$ --> create;cage=86-x15$ --> rplogit;lhs=x1;choices=arterial,rural,freeway;model: u(arterial)=dist*x7/ u(rural)=rural*one+dist*x7+cager*cage/ u(freeway)=freeway*one+dist*x7+malef*x11+cagef*cage ;fcn=dist(n);pts=200;halton$ Discrete choice and multinomial logit models Normal exit from iterations. Exit status=0. Start values obtained using MNL model Model estimated: Nov 24, 2014 at 11:46:31AM. Dependent variable Choice Iterations completed 12 Log likelihood function -97.57331 Number of parameters 6 Info. Criterion: AIC = 1.37183 Finite Sample: AIC = 1.37570 Info. Criterion: BIC = 1.49172 Info. Criterion:HQIC = 1.42054 Constants only -124.2267.21455.19592 Chi-squared[ 4] = 53.30671 Prob [ chi squared > value ] =.00000 DIST -.16731456.02997760-5.581.0000 RURAL.15641204.33257409.470.6381 CAGER.12846404.06795918 1.890.0587 FREEWAY -.06375159.72232611 -.088.9297 MALEF.55314035.63151383.876.3811 CAGEF.23491666.08450786 2.780.0054

Normal exit from iterations. Exit status=0. Random Parameters Logit Model Model estimated: Nov 24, 2014 at 11:46:34AM. Dependent variable X1 Iterations completed 13 Log likelihood function -97.17899 Number of parameters 7 Info. Criterion: AIC = 1.37985 Finite Sample: AIC = 1.38504 Info. Criterion: BIC = 1.51973 Info. Criterion:HQIC = 1.43668 Restricted log likelihood -165.8905 McFadden Pseudo R-squared.4141978 Chi squared 137.4229 Degrees of freedom 7 Prob[ChiSqd > value] =.0000000 No coefficients -165.8905.41420.40030 Constants only -124.2267.21773.19917 At start values -97.5733.00404 -.01959 Random Parameters Logit Model Replications for simulated probs. = 200 Halton sequences used for simulations ---------+Random parameters in utility functions DIST -.20497771.05791860-3.539.0004 ---------+Nonrandom parameters in utility functions RURAL.09661174.36027461.268.7886 CAGER.14332103.07504736 1.910.0562 FREEWAY -.24573833.80708702 -.304.7608 MALEF.69710541.71414306.976.3290 CAGEF.26833074.10623544 2.526.0115 ---------+Derived standard deviations of parameter distributions NsDIST.07701766.05517563 1.396.1628

Latent Class Mdoel: --> LCLOGIT;lhs=x1;choices=arterial,rural,freeway;model: u(arterial)=dist*x7/ u(rural)=rural*one+dist*x7+cager*cage/ u(freeway)=freeway*one+dist*x7+malef*x11+cagef*cage ;pts=2$ Discrete choice and multinomial logit models Normal exit from iterations. Exit status=0. Discrete choice (multinomial logit) model Model estimated: Nov 24, 2014 at 11:48:56AM. Dependent variable Choice Iterations completed 12 Log likelihood function -97.57331 Number of parameters 6 Info. Criterion: AIC = 1.37183 Finite Sample: AIC = 1.37570 Info. Criterion: BIC = 1.49172 Info. Criterion:HQIC = 1.42054 Constants only -124.2267.21455.17922 Chi-squared[ 4] = 53.30671 Prob [ chi squared > value ] =.00000 DIST 1 -.16731456.02997760-5.581.0000 RURAL 1.15641204.33257409.470.6381 CAGER 1.12846404.06795918 1.890.0587 FREEWA 1 -.06375159.72232611 -.088.9297 MALEF 1.55314035.63151383.876.3811 CAGEF 1.23491666.08450786 2.780.0054

Normal exit from iterations. Exit status=0. Latent Class Logit Model Model estimated: Nov 24, 2014 at 11:48:57AM. Dependent variable X1 Iterations completed 35 Log likelihood function -90.10460 Number of parameters 13 Info. Criterion: AIC = 1.36562 Finite Sample: AIC = 1.38322 Info. Criterion: BIC = 1.62539 Info. Criterion:HQIC = 1.47115 Restricted log likelihood -165.8905 McFadden Pseudo R-squared.4568428 Chi squared 151.5717 Degrees of freedom 13 Prob[ChiSqd > value] =.0000000 No coefficients -165.8905.45684.43241 Constants only -124.2267.27468.24205 At start values -97.5754.07656.03503 Latent Class Logit Model Number of latent classes = 2 Average Class Probabilities.680.320 ---------+Utility parameters in latent class -->> 1 DIST 1 -.55676576.27852117-1.999.0456 RURAL 1 -.42568191 1.10859747 -.384.7010 CAGER 1.33715110.24975411 1.350.1770 FREEWA 1 3.79949114 3.62111879 1.049.2941 MALEF 1 -.28126335 1.99923442 -.141.8881 CAGEF 1.61879883.34749711 1.781.0750 ---------+Utility parameters in latent class -->> 2 DIST 2 -.00490454.02979401 -.165.8692 RURAL 2.60881792.51019699 1.193.2328 CAGER 2.06272204.08942586.701.4831 FREEWA 2-1.49400393.93712737-1.594.1109 MALEF 2.95709261 1.02813651.931.3519 CAGEF 2 -.11284690.21626936 -.522.6018 ---------+Estimated latent class probabilities

PrbCls_1.67996216.02973866 22.865.0000 PrbCls_2.32003784.12925102 2.476.0133