Journal of Geographical Systems

, Volume 8, Issue 1, pp 97–108 | Cite as

Multilevel models for analyzing people’s daily movement behavior

  • Matteo Bottai
  • Nicola Salvati
  • Nicola Orsini
Original paper


A survey on the daily movement behavior of the people residing in the territory of the Municipality of Pisa, Italy, was carried out in October 2002. This work is aimed at modeling the distance covered and the number of trips taken in a day as functions of several individual characteristics. In order to take the potential intra-family and intra-area correlation of the observations into account, multilevel models are estimated. We use two and three level hierarchical linear and Poisson models to estimate the number of daily trips taken by an individual. Likelihood ratio tests indicate the movement behavior in 1 day is more alike for individuals within a family than for individuals from different families.


Multilevel Hierarchical Mixed-effects models Intra-family correlation Dependent data 


  1. Baccaïni B (1994) Migratory behavior and life cycles. Espace Popul Soc 3:61–74Google Scholar
  2. Baccaïni B (1997) Commuting and residential strategies in the Î le-de-France: individual behavior and spatial constraints. Environ Plan A 29:1801–1829CrossRefGoogle Scholar
  3. Bottai M (2003) Confidence regions when the Fisher information is zero. Biometrika 90(1):73–84CrossRefGoogle Scholar
  4. Bottai M, Barsotti O (1994) In: Lo spazio e la sua utilizzazione. Eds. Franco Angeli, MilanoGoogle Scholar
  5. Bottai M, Orsini N (2004) Confidence intervals for the variance component of random-effects linear models. Stata J 4(4):429–435Google Scholar
  6. Breslow NE, Clayton DG (1993) Approximate inference in generalized linear mixed models. J Am Stat Assoc 88:9–25CrossRefGoogle Scholar
  7. Camstra R (1996) Commuting and gender in a lifestyle perspective. Urban Stud 33P:283–300CrossRefGoogle Scholar
  8. Casella G, Berger RL (1990) Statistical inference. Duxbury Press, BelmontGoogle Scholar
  9. Cevero R, Wu K-L (1997) Polycentrism, commuting, and residential location in the San Francisco Bay area. Environ Plan A 29:865–886PubMedCrossRefGoogle Scholar
  10. Crainiceanu CM, Ruppert D (2004) Likelihood ratio tests in linear mixed models with one variance component. J R Stat Soc B 66:165–185CrossRefGoogle Scholar
  11. Dampster AP, Laird NM, Rubin DB (1977) Maximum likelihood from incomplete data via the EM algorithm. J R Stat Soc B 39:1–38Google Scholar
  12. Johnston-Anumonwo (1992) The influence of household type and gender differences in the work trip distance. Prof Geogr 44:161–169CrossRefGoogle Scholar
  13. Madden JF (1981) Why women work closer at home. Urban Stud 18:181–194CrossRefGoogle Scholar
  14. McCulloch CE, Searle SR (2001) Generalized, linear and mixed models. Wiley, New YorkGoogle Scholar
  15. Rabe-Hesketh S, Pikles A, Skrondal A (2002) GLLAMM manual. Technical Report 2001/01, Department of Biostatistics and Computing, Institute of Psychiatry, King’s College, LondonGoogle Scholar
  16. Raudeumbush SW, Bryk AS (2002) Hierarchical linear models. Sage, LondonGoogle Scholar
  17. Ruppert D, Wand MP, Carroll RJ (2003) Semiparametric regression. Cambridge University Press, CambridgeGoogle Scholar
  18. Searle SR, Casella G, McCullcoh CE (1992) Variance components. Wiley, New YorkGoogle Scholar
  19. Self SG, Liang KY (1987) Properties of maximum likelihood estimators and likelihood ratio test under nonstandard conditions. J Am Stat Assoc 82:605–610CrossRefGoogle Scholar
  20. Snijders TAB, Bosker RJ (1999) Multilevel analysis. Sage, LondonGoogle Scholar
  21. Tkocz Z, Kristensen G (1994) Commuting distances and gender: a spatial urban model. Geogr Analy 26:1–14CrossRefGoogle Scholar
  22. Vickerman RW (1984) Urban and regional change, migration and commuting—the dynamics of workplace, residence and transport choice. Urban Stud 21:15–29CrossRefGoogle Scholar
  23. White MJ (1988) Location choice and commuting behavior in cities with decentralized employment. J Urban Econ 24:129–152CrossRefGoogle Scholar

Copyright information

© Springer-Verlag 2006

Authors and Affiliations

  1. 1.Department of Epidemiology and BiostatisticsUniversity of South CarolinaColumbiaUSA
  2. 2.Dipartimento di Statistica e Matematica Applicata all’EconomiaUniversity of PisaPisaItaly
  3. 3.Institute of Environmental MedicineKarolinska InstitutetStockholmSweden

Personalised recommendations