Imbedding linear regressions in models for factor crossing

Carla Santos, Célia Nunes, Cristina Dias, Maria Varadinov, João T. Mexia

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Given u factors with J1, ⋯, Ju levels we are led to test their effects and interactions. For this we consider an orthogonal partition of Rn, with n=πl=1uJl, in subspaces associated with the sets of factors. The space corresponding to the set C will have density g(C)=πlϵC(Jl-1) so that g({1, ⋯, u}) will be much larger than the other number of degrees of freedom when Jl > 2, l = 1, ⋯, u This fact may be used to enrich these models imbedding in them linear regressions.

Original languageEnglish
Title of host publicationInternational Conference of Computational Methods in Sciences and Engineering 2016, ICCMSE 2016
PublisherAmerican Institute of Physics Inc.
Volume1790
ISBN (Electronic)9780735414549
DOIs
Publication statusPublished - 6 Dec 2016
EventInternational Conference of Computational Methods in Sciences and Engineering 2016, ICCMSE 2016 - Athens, Greece
Duration: 17 Mar 201620 Mar 2016

Conference

ConferenceInternational Conference of Computational Methods in Sciences and Engineering 2016, ICCMSE 2016
CountryGreece
CityAthens
Period17/03/1620/03/16

Keywords

  • Crossing
  • Effects and Interactions
  • Orthogonal partition

Fingerprint Dive into the research topics of 'Imbedding linear regressions in models for factor crossing'. Together they form a unique fingerprint.

  • Cite this

    Santos, C., Nunes, C., Dias, C., Varadinov, M., & Mexia, J. T. (2016). Imbedding linear regressions in models for factor crossing. In International Conference of Computational Methods in Sciences and Engineering 2016, ICCMSE 2016 (Vol. 1790). [140005] American Institute of Physics Inc.. https://doi.org/10.1063/1.4968734