User:White Silver

Variance ratio test

The variance ratio test is a statistical test that examines whether a time series follows a random walk. The variance ratio test has been proposed by Cochrane (1988) and Lo and MacKinlay (1988).

Motivation and mathematical background

Consider a time series of the following form

Y_t = μ + Y_t-1 + U _t

U_t = ∑ψ_jε_t-j = Ψ(L)ε_t

Where Y_t is the natural logarithm of the value of the time series at time t (e.g. the daily close of the S&P 500 index), μ is the drift of the time series, Y_t-1 is the value of the time series at time t-1 (e.g. the daily close of the S&P 500 on the previous day), U_t is a disturbance term governed by a moving average process with coefficients ψ_j (with the subscript j running from 0 to ∞), Ψ(L) is thus an infinite polynomial in the lag operator, and ε_t is a covariance stationary white noise process with the following three attributes: 1. E(ε_t)=0, 2. var(ε_t)=σ², and 3. cov(ε_t, ε_t-j)=γ_j.