No label defined (Q3750858)
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 3988542
| Language | Label | Description | Also known as |
|---|---|---|---|
| default for all languages | No label defined |
||
| English | No label defined |
scientific article; zbMATH DE number 3988542 |
Statements
1985
0 references
chi-square test
0 references
regular, homogeneous and stationary Markov chain with two states
0 references
limit behavior
0 references
rate of convergence
0 references
0.8436058163642883
0 references
[For the entire collection see Zbl 0594.00015.] Consider a regular, homogeneous and stationary Markov chain with two states denoted by \(\{\) 0,1\(\}\). Let \(P=(p_{ij})\), \(p_{ij}>0\), i,j\(\in \{0,1\}\) and \(p^ 0=(p^ 0_ 0,p^ 0_ 1)\) be the transition matrix and the initial distribution of the chain. For given \(n+1\) consecutive observations \(X_ 1,X_ 2,...,X_{n+1}\) of the chain, denote by \(n_{ij}\) the number of transitions of the kind (i,j) where i,j\(\in \{0,1\}\) and by \(n=\sum_{i,j}n_{ij}\) the total number of these transitions. The quantity \[ \chi^ 2_ n=\sum_{i,j}(n_{ij}-n_ ip_{ij})^ 2/(np_ ip_{ij}) \] plays an important role in various stochastic problems. Recall that in the case of independent observations \(\chi^ 2_ n\) converges in distribution to a random variable which is \(\chi^ 2\)-distributed with 2 degrees of freedom. Note however that the Markovian case treated in this paper is much more complicated. The authors have found the limit behavior of the probability \(P\{\chi^ 2_ n\leq r^ 2\}\) when \(r=O(1)\) as \(n\to \infty\) as well as the rate of convergence. This result is applied successfully to some related statistical problems.
0 references