Analysis of Repeated Measurement Problem in SP data

SP 데이터의 Repeated Measurement Problem 분석

  • CHO, Hye-Jin (Senior Researcher Korea Institute of construction Technology)
  • Published : 2002.02.01

Abstract

One of the advantages of SP methods is the possibility of getting a number of responses from each respondent. However, when the repeated observations from each respondent are analysed by applying the simple modeling method, a potential problem is created because of upbiased significance due to the repeated observation from each respondent. This study uses a variety of approaches to explore this issue and to test the robustness of the simple model estimates. Among several different approaches, the Jackknife method and Kocurs method were applied. The Jackknife method was implemented using a program JACKKNIFE. The model estimate results of Jackknife method and Kocurs method were compared with those of the uncorrected estimates in order to test whether there was repeated measurement problem or not and the extent to which this problem affected the model estimates. The standard errors between the uncorrected model estimates and Jackknife estimates were also compared. The results reveals that the t-ratios of Kocurs are much lower than those of the uncorrected method and Jackknife estimates, indicating that Kocurs method underestimates the significance of the coefficients. Jackknife method produced the almost same coefficients as those of the uncorrected model but the lower t-ratios. These results indicate that the coefficients of the uncorrected method are accurate but that their significance are somewhat overestimated. In this study. 1 concluded that the repeated measurement Problem did exist in our data, but that it did not affect the model estimation results significantly. It is recommended that such a test should become a standard procedure. If it turns out that the analysis based on the simple uncorrected method are influenced by the repeated measurement problem. it should be corrected.

SP 방법의 장점중 하나는 개인의 응답자로부터 다수의 응답(Repeated observations)을 받을 수 있는 점이다. 그러나 이렇게 얻은 개인의 다중응답을 각각의 응답이 독립적이라는 것을 가정하고 있는 단순한 모델링방법을 이용해서 분석하면 t값이 상향되는 문제를 가져올 수 있다. 이러한 문제를 다중응답의 문제(Reapeated Measurement Problem)라고 한다 본 연구는 다양한 접근을 통해서 이러한 다중응답의 문제(Reapeated Measurement Problem)를 다루고, 단순한 모델링방법을 통한 모델추정치가 신뢰할 만한가를 검증하도록 한다. 다양한 접근중에서 본 연구에서는 Jackknife와 Kocurs 방법을 적용하였다. Jackknife 방법은 JACKKNIFE란 소프트웨어를 사용해서 분석하였다. Jackknife 추정치와 Kocurs 추정치를 단순한 모델링기법을 적용한 모델추정치와 비교하여서 다중응답의 문제(Repeated Measurement Problem)의 발생여부와 모델추정치에 어느정도 영향을 미치는 지를 분석하였다. 단순모델링의 추정치와 Jackknife 추정치의 표준오차도 비교하였다. 결과를 요약하면. Kocurs방법의 추정치의 t값은 단순한 모델링기법과 Jackknife 추정치의 t값보다 매우 낮게 나타났으며, 이는 Kocurs방법은 파라미터의 유의성을 지나치게 하향추정하는 것을 의미한다. Jackknife 방법의 추정치는 단순한 모델링기법의 추정치와 계수값은 거의 유사하나 t값이 다소 적은 것으로 나타났다. 이러한 결과들은 단순한 모델링기법의 계수값은 정확하나 유의정도가 다소 과장되는 것을 의미한다. 결론적으로 본 연구에서 사용한 데이터에는 다중응답의 문제가 존재하나 모델추정 티에 유의하게 영향을 미치지 않는 것으로 나타났다 SP 방법을 사용한 데이터를 분석할 경우 다중응답의 문제(Repeated Measurement Problem)를 분석하는 절차를 반드시 수행하여야 한다. 만약 단순한 모델링기법의 추정치가 다중응답의 문제로 인해서 영향을 받았다면 반드시 모델링 추정치를 보정하여야 한다.

Keywords

References

  1. The Statistician v.24 no.2 The Jockknife-Toy,Tool or Two-edged Weapon? Bissell A.J.;Ferguson,R.A. https://doi.org/10.2307/2987663
  2. UTSG Conference The effects of combination of advanced traveller information systems with congestion pricing :Pilot stated perference survey Cho.H.J.
  3. PhD Thesis.Institute for Transprot Studies Diver's Responses to Variable Road User Charges and Traffic Information Cho, H.J.
  4. PTRC.24th.Sept Eliminating Bias due ro the Repeated Measurements Problem in SP data Cirillo.C.;Darly,A.;Lindveld,K.
  5. Monographs on Statistics and Appliced Provability An introduction to the bootstrap Efton,B.;Tivshirani,R.J.
  6. Analusing Demand for Rail Travel.Avebury(Chapter 4) Disaggregate Methods Fowkes,A.S.;Wardman.M.;Nash.C.A.;
  7. Working Paper. Institute for Transprot Studies The Value of Business Travel Time Savins Fowkes,A.S;Marks.P.;Nash,C.A.
  8. Institute for Transprot Studies.Working Paper. The Distributional Impact of Various Road Schems for London Fowkes,A.S;Miline,D.S.;Nash,C.A.;May,A.D.
  9. Proceeding of the WCTR conference Analysing Repeated Measurement Problems in SP Data Modelling Kim.K.S.
  10. Urban Mass Transportation Administration Report(no.UNITA-NH11-C001-82-1) Guide To Forecasting Travel Demand with Direct Utility Assessment Kocur,G.T.;Adler,T.;Hyman.W.;Aunnet,B
  11. Modelling Route and Multimodal Chocies with Revealed And Stated Preference Data Ortuzar,J.D.
  12. Journey of Transport Economics and Policy Stated Chocie Experiments with Repeated Observations Ouwesloot,H.;Retveld.P.
  13. Stated Preference Techniques-A guide to Practics Pearmain,D.;Kroes.E.
  14. The Jackknife and Bootstop(Springer Series in Statistics) Shao;Tu
  15. Institute for Transport Studies.Working Stated Preference Analysis Of Driver Route Choice Reaction To VArinable Message Sign Information Wardman,M.;Bonsall,P.W.;Shires,J.
  16. Introductory Statistics For Business and Economics(Forth Edition) Wonnacott,T.H.;Wommacott,R.J.
  17. Transpn.Res.-B v.31 no.3 Multinomial Probit With Structured Covaraince For Route Choice Behaviour Yai.T;Iwakura,S.;Morichi.S. https://doi.org/10.1016/S0191-2615(96)00025-2