ABSTRACT
We derive the exact distribution of the maximum likelihood estimator of the mean reversion parameter (κ) in the Ornstein–Uhlenbeck process using numerical integration through analytical evaluation of a joint characteristic function. Different scenarios are considered: known or unknown drift term, fixed or random start-up value, and zero or positive κ. Monte Carlo results demonstrate the remarkably reliable performance of our exact approach across all the scenarios. In comparison, misleading results may arise under the asymptotic distributions, including the advocated infill asymptotic distribution, which performs poorly in the tails when there is no intercept in the regression and the starting value of the process is nonzero.
Acknowledgments
We are grateful to Peter Phillips and three anonymous referees for their constructive feedback. We also thank Raymond Kan for his discussion of the numerical algorithm presented in this paper.
Notes
1Since we are interested in studying the finite-sample properties of the initial condition x0 matters and we include it in the estimation procedure. This stands in contrast to the convention of Hurwicz (Citation1950). For the case of known μ, if μ≠0, one can simply define and work with yi.
2For 0<ϕ≤1, asymptotically, since is consistent. In other words, as in (2.5) is always well defined asymptotically, and so is its asymptotic distribution. We may define an inequality constrained estimator of κ following the spirit of Judge and Takayama (Citation1966), but this approach is not pursued in this paper.
3Note that (2.6) holds regardless of the distributional assumption.
4If we discard x0 in formulating , then the Imhof (Citation1961) technique is still applicable, as we can define in terms of quadratic forms in the random vector
5As emphasized in Zhou and Yu (2015), under the infill asymptotic regime, the same notation is used for any value of κ, demonstrating that the infill distribution is continuous in κ. When κ = 0, the terms , , and should be interpreted as their limiting values as κ→0.
6http://www.krannert.purdue.edu/faculty/ybao/OU.zip. The file names are self-suggestive.
7In simulating the asymptotic (nonnormal) results, we used a sample size of 5,000 and 100,000 replications to approximate the integrals involving the Brownian motion by the discrete Riemann sums.
8More specifically, let denote the integrand function in question with t∈[l,u]. quadgk requires the integrand function to accept a vector and returns a vector of output. Let and denote , and The algorithm is as follows: (i) Start with t1 and set Set k = 0. (ii) Beginning with t2, if ai<0, ai−1<0, and set k = k+1; otherwise, k is unchanged. Set where
9Phillips (Citation1980) provided analytical characteristic functions involving the more general cases of linear and quadratic forms.
10 is independent of μ because , , and .
11This is different from the case when no intercept is included in the AR(1) model.