IDEAS home Printed from https://ideas.repec.org/p/pra/mprapa/77790.html
   My bibliography  Save this paper

Evaluating Neural Spatial Interaction Modelling by Bootstrapping

Author

Listed:
  • Fischer, Manfred M.
  • Reismann, Martin

Abstract

This paper exposes problems of the commonly used technique of splitting the available data in neural spatial interaction modelling into training, validation, and test sets that are held fixed and warns about drawing too strong conclusions from such static splits. Using a bootstrapping procedure, we compare the uncertainty in the solution stemming from the data splitting with model specific uncertainties such as parameter initialization. Utilizing the Austrian interregional telecommunication traffic data and the differential evolution method for solving the parameter estimation task for a fixed topology of the network model [ i.e. J = 9] this paper illustrates that the variation due to different resamplings is significantly larger than the variation due to different parameter initializations. This result implies that it is important to not over-interpret a model, estimated on one specific static split of the data.

Suggested Citation

  • Fischer, Manfred M. & Reismann, Martin, 2000. "Evaluating Neural Spatial Interaction Modelling by Bootstrapping," MPRA Paper 77790, University Library of Munich, Germany.
    • Handle: RePEc:pra:mprapa:77790
    as

    Download full text from publisher

    File URL: https://mpra.ub.uni-muenchen.de/77790/1/MPRA_paper_77790.pdf
    File Function: original version
    Download Restriction: no
    ---><---

    Other versions of this item:

    References listed on IDEAS

    as
    1. Manfred M. Fischer & Katerina Hlavácková-Schindler & Martin Reismann, 1999. "articles: A global search procedure for parameter estimation in neural spatial interaction modelling," Papers in Regional Science, Springer;Regional Science Association International, vol. 78(2), pages 119-134.
    2. M M Fischer, 1998. "Computational Neural Networks: A New Paradigm for Spatial Analysis," Environment and Planning A, , vol. 30(10), pages 1873-1891, October.
    3. Fischer, Manfred M. & Gopal, Sucharita, 1994. "Artificial Neural Networks. A New Approach to Modelling Interregional Telecommunication Flows," MPRA Paper 77822, University Library of Munich, Germany.
    4. S Openshaw, 1998. "Neural Network, Genetic, and Fuzzy Logic Models of Spatial Interaction," Environment and Planning A, , vol. 30(10), pages 1857-1872, October.
    Full references (including those not matched with items on IDEAS)

    Citations

    Citations are extracted by the CitEc Project, subscribe to its RSS feed for this item.
    as


    Cited by:

    1. Fischer, Manfred M. & Reismann, Martin, 2002. "A Methodology for Neural Spatial Interaction Modeling," MPRA Paper 77794, University Library of Munich, Germany.
    2. Manfred M. Fischer, 2009. "Principles of Neural Spatial Interaction Modeling," Advances in Spatial Science, in: Michael Sonis & Geoffrey J. D. Hewings (ed.), Tool Kits in Regional Science, chapter 8, pages 199-214, Springer.
    3. Giuseppe Bruno & Andrea Genovese, 2012. "A Spatial Interaction Model for the Representation of the Mobility of University Students on the Italian Territory," Networks and Spatial Economics, Springer, vol. 12(1), pages 41-57, March.
    4. Longhi, Simonetta & Nijkamp, Peter & Reggiani, Aura & Blien, Uwe, 2002. "Forecasting regional labour markets in Germany: an evaluation of the performance of neural network analysis," ERSA conference papers ersa02p117, European Regional Science Association.

    Most related items

    These are the items that most often cite the same works as this one and are cited by the same works as this one.
    1. P Rees & I Turton, 1998. "Guest Editorial," Environment and Planning A, , vol. 30(10), pages 1835-1838, October.
    2. Fischer, Manfred M., 2006. "Neural Networks. A General Framework for Non-Linear Function Approximation," MPRA Paper 77776, University Library of Munich, Germany.
    3. Manfred M. Fischer, 2009. "Principles of Neural Spatial Interaction Modeling," Advances in Spatial Science, in: Michael Sonis & Geoffrey J. D. Hewings (ed.), Tool Kits in Regional Science, chapter 8, pages 199-214, Springer.
    4. Nijkamp, Peter & Reggiani, Aura & Tsang, Wai Fai, 2004. "Comparative modelling of interregional transport flows: Applications to multimodal European freight transport," European Journal of Operational Research, Elsevier, vol. 155(3), pages 584-602, June.
    5. Nijkamp, Peter & Reggiani, Aura & Sabella, E., 1999. "A comparison of the performance of Discrete Choice Models and Biocomputing Models in Transport Systems Analysis," ERSA conference papers ersa99pa037, European Regional Science Association.
    6. Roberto Patuelli & Aura Reggiani & Peter Nijkamp & Norbert Schanne, 2011. "Neural networks for regional employment forecasts: are the parameters relevant?," Journal of Geographical Systems, Springer, vol. 13(1), pages 67-85, March.
    7. Aura Reggiani & Peter Nijkamp & Enrico Sabella, 1998. "Evolutionary algorithms: Overview and applications to European transport," ERSA conference papers ersa98p412, European Regional Science Association.
    8. Xia Li & Anthony Gar-On Yeh, 2001. "Calibration of Cellular Automata by Using Neural Networks for the Simulation of Complex Urban Systems," Environment and Planning A, , vol. 33(8), pages 1445-1462, August.
    9. Tom Kauko, 2005. "Using the Self-Organising Map to Identify Regularities across Country-Specific Housing-Market Contexts," Environment and Planning B, , vol. 32(1), pages 89-110, February.
    10. Nijkamp, Peter & Reggiani, Aura & Bolis, Simona, 1997. "European freight transport and the environment: empirical applications and scenarios," Serie Research Memoranda 0033, VU University Amsterdam, Faculty of Economics, Business Administration and Econometrics.
    11. Fischer, Manfred M. & Reismann, Martin, 2002. "A Methodology for Neural Spatial Interaction Modeling," MPRA Paper 77794, University Library of Munich, Germany.
    12. Ningchuan Xiao & David A Bennett & Marc P Armstrong, 2002. "Using Evolutionary Algorithms to Generate Alternatives for Multiobjective Site-Search Problems," Environment and Planning A, , vol. 34(4), pages 639-656, April.
    13. Tom Kauko, 2004. "A Comparative Perspective on Urban Spatial Housing Market Structure: Some More Evidence of Local Sub-markets Based on a Neural Network Classification of Amsterdam," Urban Studies, Urban Studies Journal Limited, vol. 41(13), pages 2555-2579, December.
    14. Roberto Patuelli & Peter Nijkamp & Simonetta Longhi & Aura Reggiani, 2008. "Neural Networks and Genetic Algorithms as Forecasting Tools: A Case Study on German Regions," Environment and Planning B, , vol. 35(4), pages 701-722, August.
    15. Fischer, Manfred M. & Openshaw, Stan, 1995. "A Framework for Research on Spatial Analysis Relevant to Geo-Statistical Informations Systems in Europe," MPRA Paper 77814, University Library of Munich, Germany.
    16. B Rubenstein-Montano & I Zandi, 1999. "Application of a Genetic Algorithm to Policy Planning: The Case of Solid Waste," Environment and Planning B, , vol. 26(6), pages 893-907, December.
    17. Javier Rubio-Herrero & Jesús Muñuzuri, 2021. "Indirect estimation of interregional freight flows with a real-valued genetic algorithm," Transportation, Springer, vol. 48(1), pages 257-282, February.
    18. Reggiani, Aura & Nijkamp, Peter & Sabella, Enrico, 2001. "New advances in spatial network modelling: Towards evolutionary algorithms," European Journal of Operational Research, Elsevier, vol. 128(2), pages 385-401, January.
    19. Karima Kourtit & Daniel Arribas-Bel & Peter Nijkamp, 2012. "High performers in complex spatial systems: a self-organizing mapping approach with reference to The Netherlands," The Annals of Regional Science, Springer;Western Regional Science Association, vol. 48(2), pages 501-527, April.
    20. Roberto Patuelli & Aura Reggiani & Peter Nijkamp & Uwe Blien, 2006. "New Neural Network Methods for Forecasting Regional Employment: an Analysis of German Labour Markets," Spatial Economic Analysis, Taylor & Francis Journals, vol. 1(1), pages 7-30.

    More about this item

    Keywords

    Neural spatial interaction modelling; model evaluation; bootstrapping; interregional telecommunications;
    All these keywords.

    JEL classification:

    • C21 - Mathematical and Quantitative Methods - - Single Equation Models; Single Variables - - - Cross-Sectional Models; Spatial Models; Treatment Effect Models
    • C45 - Mathematical and Quantitative Methods - - Econometric and Statistical Methods: Special Topics - - - Neural Networks and Related Topics

    Statistics

    Access and download statistics

    Corrections

    All material on this site has been provided by the respective publishers and authors. You can help correct errors and omissions. When requesting a correction, please mention this item's handle: RePEc:pra:mprapa:77790. See general information about how to correct material in RePEc.

    If you have authored this item and are not yet registered with RePEc, we encourage you to do it here. This allows to link your profile to this item. It also allows you to accept potential citations to this item that we are uncertain about.

    If CitEc recognized a bibliographic reference but did not link an item in RePEc to it, you can help with this form .

    If you know of missing items citing this one, you can help us creating those links by adding the relevant references in the same way as above, for each refering item. If you are a registered author of this item, you may also want to check the "citations" tab in your RePEc Author Service profile, as there may be some citations waiting for confirmation.

    For technical questions regarding this item, or to correct its authors, title, abstract, bibliographic or download information, contact: Joachim Winter (email available below). General contact details of provider: https://edirc.repec.org/data/vfmunde.html .

    Please note that corrections may take a couple of weeks to filter through the various RePEc services.

    IDEAS is a RePEc service. RePEc uses bibliographic data supplied by the respective publishers.