SEM State Space Modeling of Panel Data in Discrete and Continuous Time and its Relationship to Traditional State Space Modeling

  • Johan Oud
Part of the Mathematical Modelling: Theory and Applications book series (MMTA, volume 19)

Abstract

The state space approach, introduced by Kalman (1960) and Zadeh and Desoer (1963), distinguishes modern systems and control theory (Cames, 1988; Hannan & Deistler, 1988; Polderman & Willems, 1998) from the classical theory. Its application field, starting from space technology in the 1960s, has extended considerably since (Shumway & Stoffer, 2000). The state space model covers an extremely general class of dynamic models. In fact, all nonanticipative models (models with no causal arrows heading backward in time) can be represented in state space form. For example, both the Box-Jenkins ARMA model and the extended ARMAX model, which adds exogenous or input variables to the ARMA model, are easily formulated as special cases of the state space model (Cames, 1988; Deistler, 1985; Ljung, 1985). The state space model covers also longitudinal latent factor and path analysis models and allows the optimal estimation of the latent states or factor scores (Oud, van den Bercken & Essers, 1990; Oud, Jansen, van Leeuwe, Aarnoutse, & Voeten, 1999). Latent state estimation is performed by two important results of the state space approach: the Kalman filter and smoother (Kalman, 1960; Rauch, Tung, & Striebel, 1965).

Keywords

Covariance Income Marin 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. Anderson, T.W. (1958). An introduction to multivariate statistical analysis. New York: Wiley.Google Scholar
  2. Arbuckle, J.L., & Wothke, W. (1999). AMOS 4.0 user’s guide. Chicago: Smallwaters.Google Scholar
  3. Arnold, L. (1974). Stochastic differential equations. New York: Wiley.Google Scholar
  4. Baltagi, B.H. (1995). Econometric analysis of panel data. Chichester: Wiley.Google Scholar
  5. Bergstrom, A.R. (1984). Continuous time stochastic models and issues of aggregation over time. In Z. Griliches & M.D. Intriligator (Eds.), Handbook of econometrics: Vol. 2. (pp. 1145–1212). Amsterdam: North-Holland.CrossRefGoogle Scholar
  6. Bergstrom, A.R. (1988). The history of continuous-time econometric models. Econometric Theory, 4, 365–383.CrossRefGoogle Scholar
  7. Browne, M.W., & Arminger, G. (1985). Specification and estimation of mean-and covariance-structure models. In G. Arminger, C.C. Clogg, & M.E. Sobel (Eds.), Handbook of statistical modeling for the social and behavioral sciences(pp. 185–249). New York: Plenum Press.Google Scholar
  8. Caines, P.E. (1988). Linear stochastic systems. New York: Wiley.Google Scholar
  9. Caines, P.E., & Rissanen, J. (1974). Maximum likelihood estimation of pa­rameters in multivariate Gaussian processes. IEEE Transactions on Information Theory, 20, 102–104.CrossRefGoogle Scholar
  10. Deistler, M. (1985). General structure and parametrization of ARMA and state-space systems and its relation to statistical problems. In E.J. Hannan, P.R. Krishnaiah, & M.M. Rao (Eds.), Handbook of statistics: Vol. 5. Time series in the time domain (pp. 257–277). Amsterdam: North-Holland.Google Scholar
  11. Dembo, A., & Zeitouni, O. (1986). Parameter estimation of partially observed continuous time stochastic processes via the EM algorithm. Stochas­tic Processes and their Applications, 23, 91–113.CrossRefGoogle Scholar
  12. Desoer, C.A. (1970). Notes for a second course on linear systems. New York: Van Nostrand Reinhold.Google Scholar
  13. Goodrich, R.L., & Caines, P.E. (1979). Linear system identification from non-stationary cross-sectional data. IEEE Transactions on Automatic Control, 1979,403–411.CrossRefGoogle Scholar
  14. Gandolfo, G. (1993). Continuous-time econometrics has come of age. In G. Gandolfo (Ed.), Continuous time econometrics (pp. 1–11). London: Chapman & Hall.CrossRefGoogle Scholar
  15. Hamerle, A., Nagl, W., & Singer, H., (1991). Problems with the estimation of stochastic differential equations using structural equation models. Journal of Mathematical Sociology, 16, 201–220.CrossRefGoogle Scholar
  16. Hamerle, A., Singer, H., & Nagl, W. (1993). Identification and estimation of continuous time dynamic systems with exogenous variables using panel data. Econometric Theory, 9, 283–295.CrossRefGoogle Scholar
  17. Hannan, E.J., & Deistler, M. (1988). The statistical theory of linear systems. New York: Wiley.Google Scholar
  18. Hertzog, C, & Nesselroade, J.R. (1987). Beyond autoregressive models: Some implications of the trait-state distinction for the structural mod­eling of developmental change. Child Development, 58, 93–109.PubMedCrossRefGoogle Scholar
  19. Hsiao, C. (1986). Analysis of panel data. Cambridge: Cambridge University Press.Google Scholar
  20. Jansen, R.A.R.G., & Oud, J.H.L. (1995). Longitudinal LISREL model esti­mation from incomplete panel data using the EM algorithm and the Kalman smoother. Statistica Neerlandica, 49, 362–377.CrossRefGoogle Scholar
  21. Jazwinski, A.H. (1970). Stochastic processes and filtering theory. New York: Academic Press.Google Scholar
  22. Jones, R.H. (1985). Time series analysis with unequally spaced data. In E.J. Hannan, P.R. Krishnaiah, & M.M. Rao (Eds.), Handbook of statistics: Vol. 5. Time series in the time domain (pp. 157–177). Amsterdam: North-Holland.Google Scholar
  23. Joreskog, K.G. (1973). A general method for estimating a structural equation system. In A.S. Goldberger & O.D. Duncan (Eds.), Structural equation models in the social sciences (pp. 85–112). New YorkGoogle Scholar
  24. Joreskog, K.G., & Sorbom, D. (1996). LISREL 8: User’s reference guide. Chicago: Scientific Software International.Google Scholar
  25. Kalman, R.E. (1960). A new approach to linear filtering and prediction problems. Journal of Basic Engineering(Trans. ASME, ser. D), 82, 35–45.CrossRefGoogle Scholar
  26. Little, D.J.A., & Rubin, D.B. (1987). Statistical analysis with missing data. New York: Wiley.Google Scholar
  27. Ljung, L. (1985). Estimation of parameters in dynamical systems. In E.J. Hannan, RR. Krishnaiah, & M.M. Rao (Eds.), Handbook of statistics: Vol. 5. Time series in the time domain (pp. 189–211). Amsterdam: North-Holland.Google Scholar
  28. MacCallum, R., & Ashby, EG. (1986). Relationships between linear systems theory and covariance structure modeling. Journal of Mathematical Psychology, 30, 1–27.CrossRefGoogle Scholar
  29. Meditch, J.S. (1969). Stochastic optimal linear estimation and control. New York: McGraw-Hill.Google Scholar
  30. Mehra, R.K. (1971). Identification of stochastic linear dynamic systems using Kalman filter representation. AlAA Journal23, 28–31.Google Scholar
  31. Molenaar, RC.M. (1985). A dynamic factor model for the analysis of multi­variate time series. Psychometrika t 50., 181–202.CrossRefGoogle Scholar
  32. Molenaar, RC.M., de Gooijer, J.G., & Schmitz, B. (1992). Dynamic factor analysis of nonstationary multivariate time series. Psychometrika,57, 333–349.CrossRefGoogle Scholar
  33. Neale, M.C. (2000). Individual fit, heterogeneity, and missing data in multi-group structural equation modeling. In T.D. Little, K.U. Schnabel, & J. Baumert (Eds.), Modeling longitudinal and multilevel data(pp. 219–240). Mahwah NJ: Lawrence Erlbaum.Google Scholar
  34. Neale, M.C, Boker, S.M., Xie, G., & Maes, H.H. (1999) Mx: Statistical Modeling(5th ed.). Richmond VA: Department of Psychiatry.Google Scholar
  35. Oud, J.H.L. (1978). Syteem-methodologie in sociaal-wetenschappelijk onder-zoek [Systems methodology in social science research]. Doctoral dissertation. Nijmegen, The Netherlands: Alfa.Google Scholar
  36. Oud, J.H.L. (2001). Quasi-longitudinal designs in SEM state space modeling. Statistica Neerlandica t 55., 200–220.CrossRefGoogle Scholar
  37. Oud, J.H.L. (2002). Continuous time modeling of the cross-lagged panel de­sign. Kwantitatieve Methoden, 23(69), 1–26.Google Scholar
  38. Oud, J.H.L., & Jansen, R.A.R.G. (1995). An ARMA extension of the longitudinal LISREL model for LISKAL. In I. Parchev (Ed.), Multivariateanalysis in the behavioral sciences: philosophic to technical, (pp. 49–69). Sofia: “Prof. Marin Drinov” Academic Publishing House.Google Scholar
  39. Oud, J.H.L., & Jansen, R.A.R.G. (1996). Nonstationary longitudinal LIS-REL model estimation from incomplete panel data using EM and the Kalman smoother. In U. Engel & J. Reinecke (Eds.), Analysis of change: Advanced techniques in panel data analysis (pp. 135–159). Berlin: Walter de Gruyter.Google Scholar
  40. Oud, J.H.L., & Jansen, R.A.R.G. (2000). Continuous time state space modeling of panel data by means of SEM. Psychometrika, 65, 199–215.CrossRefGoogle Scholar
  41. Oud, J.H.L., Jansen, R.A.R.G., van Leeuwe, J.F.J., Aarnoutse, C.A.J., & Voeten, M.J.M. (1999). Monitoring pupil development by means of the Kalman filter and smoother based upon SEM state space modeling. Learning and Individual Differences, 11, 121–136.CrossRefGoogle Scholar
  42. Oud, J.H.L., van den Bercken, J.H.L., & Essers, R.J. (1986). Longitudinal factor scores estimation using the Kalman filter. Kwantitatieve Methoden, 7(20), 109–130.Google Scholar
  43. Oud, J.H.L., van den Bercken, J.H.L., & Essers, R.J. (1990). Longitudinal factor scores estimation using the Kalman filter. Applied Psychological Measurement, 14, 395–418.CrossRefGoogle Scholar
  44. Oud, J.H.L., van Leeuwe, J.F.J., & Jansen, R.A.R.G. (1993). Kalman filtering in discrete and continuous time based on longitudinal LISREL mod­els. In J.H.L. Oud & A.W. van Blokland-Vogelesang (Eds.), Advances in longitudinal and multivariate analysis in the behavioral sciences (pp. 3–26). Nijmegen: ITS.Google Scholar
  45. Phadke, M.S., & Wu, S.M. (1974). Modeling of continuous stochastic processes from discrete observations with applications to sunspot data. Journal of the American Statistical Association, 69, 325–329.CrossRefGoogle Scholar
  46. Phillips, P.C.B. (1993). The ET Interview: A.R. Bergstrom. In P.C.B. Phillips (Ed.), Models, methods, and applications of econometrics (pp. 12–31). Cambridge MA: Blackwell.Google Scholar
  47. Polderman, J.W., & Willems, J.C. (1998). Introduction to mathematical systems theory: A behavioral approach. Mooresville IN: Scientific Software.CrossRefGoogle Scholar
  48. Rauch, H.E., Tung, E, & Striebel, C.T. (1965). Maximum likelihood estimates of linear dynamic systems. AIAA Journal, 3, 1445–1450.CrossRefGoogle Scholar
  49. Ruymgaart, Pa. & Soong, T.T. (1985). Mathematics of Kalman-Bucy filtering. Berlin: Springer.CrossRefGoogle Scholar
  50. Schweppe, F. (1965). Evaluation of likelihood functions for Gaussian signals. IEEE Transactions on Information Theory, 11, 61–70.CrossRefGoogle Scholar
  51. Shumway, R.H., & Stoffer, D.S. (1982). An approach to time series smoothing and forecasting using the EM algorithm. Journal of Time Series Analysis, 3, 253–264.CrossRefGoogle Scholar
  52. Shumway, R.H., & Stoffer, D.S. (2000). Time series analysis and its applica­tions. New York: Springer.CrossRefGoogle Scholar
  53. Singer, H. (1990). Parameterschdtzung in zeitkontinuierlichen dynamischen Systemen. Konstanz: Hartung-Gorre.Google Scholar
  54. Singer, H. (1993). Continuous-time dynamical systems with sampled data, errors of measurement and unobserved components. Journal of Time Series Analysis, 14, 527–545.CrossRefGoogle Scholar
  55. Wothke, W. (2000). Longitudinal and multigroup modeling with missing data. In T.D. Little, K.U. Schnabel, & J. Baumert (Eds.), Modeling longitudinal and multilevel data(pp. 219–240). Mahwah NJ: Lawrence Erlbaum.Google Scholar
  56. Young, P. (1985). Recursive identification, estimation and control. In E.J. Hannan & P.R. Krishnaiah (Eds.), Handbook of statistics, vol 5: Time series in the time domain(pp. 213–255). Amsterdam: North-Holland.Google Scholar
  57. Zadeh, L.A., & Desoer, C.A. (1963). Linear System theory: The state space approach. New York: McGraw-Hill.Google Scholar

Copyright information

© Springer Science+Business Media Dordrecht 2004

Authors and Affiliations

  • Johan Oud
    • 1
    • 2
  1. 1.Institute of Family and Child Care StudiesUniversity of NijmegenNijmegenThe Netherlands
  2. 2.Monitoring Systems ConsultNijmegenThe Netherlands

Personalised recommendations