Maximum Likelihood Estimation in the Mover-Stayer Model (Q3218964)

From MaRDI portal

!

This is the item page for this Wikibase entity, intended for internal use and editing purposes. Please use the normal view instead:

scientific article; zbMATH DE number 3885157
Language Label Description Also known as
default for all languages
No label defined
    English
    Maximum Likelihood Estimation in the Mover-Stayer Model
    scientific article; zbMATH DE number 3885157

      Statements

      0 references
      1984
      0 references
      multiwave panel data
      0 references
      mover-stayer model
      0 references
      mixture of two independent Markov chains
      0 references
      maximum likelihood estimators
      0 references
      strong consistency
      0 references
      Maximum Likelihood Estimation in the Mover-Stayer Model (English)
      0 references
      The mover-stayer model is a discrete time stochastic process \(Z_ j\), \(j=0,1,..\). that is a mixture of two independent Markov chains, one degenerate with transition probability matrix equal to the identity matrix I and the other with transition matrix \(M=\| m_{ik}\|\). The transition probabilities of Z are given by \(P[Z_ j=k| Z_ 0=i]=(1-s_ i)m_{ik}^{(j)}\) if \(i\neq k\) and \(=s_ i+(1-s_ i)m_{ii}^{(j)}\) if \(i=k\), where \(j=1,2,...\), \(1\leq i,k\leq w\), \(\| m_{ik}^{(j)}\|\) is the jth power of M, and \((s_ i,1-s_ i)\), \(0<s_ i<1\) is a mixing measure for the state \(i\in S=\{1,2,...,w\}\). The problem is to estimate the parameters of the model, i.e. the matrix M and the mixing measure \(S=\{s_ i,i\in S\}\) on the basis of n independent realizations of the process Z. The author has found explicitly the maximum likelihood estimators for M and s and has proved their strong consistency. Comparison is done with other estimators previously proposed in the literature. Numerical comparison is presented, too. Finally an application is given to testing the hypothesis about the type of the model.
      0 references

      Identifiers

      0 references
      0 references
      0 references
      0 references
      0 references
      0 references