Consider a stationary normal process ξ(t) with mean zero and the covariance function r(t). Properties of the sample functions in the neighborhood of zeros, upcrossings of very high levels, etc. have been studied by, among others, Kac and Slepian, 1959 [4] and Slepian, 1962 [11]. In this paper we shall study the sample functions near local maxima of height u, especially as u → -∞, and mainly use similar methods as [4] and [11]. Then it is necessary to analyse carefully what is meant by "near a maximum of height u." In Section 2 we derive the "ergodic" definition, i.e. the definition which is possible to interpret by the aid of relative frequencies in a single realisation. This definition has been treated previously by Leadbetter, 1966 [5], and it turns out to be related to Kac and Slepian's horizontal window definition. In Section 3 we give a representation of ξ(t) near a maximum as the difference between a non-stationary normal process and a deterministic process, and in Section 4 we examine these processes as u → -∞. We have then to distinguish between two cases. A: Regular case. r(t) = 1 -λ2t2/2 + λ4 t4/4! - λ6 t6/6! + o(t6) as t → 0, where the positive λ2k are the spectral moments. Then it is proved that if ξ(t) has a maximum of height u at t = 0 then, as u → -∞, egin{align*} (lambda_2lambda_6 - lambda_4^2)(lambda_4 - lambda_2^2)^{-1}{xi((lambda_2lambda_6 - lambda_4^2)^{-frac{1}{2}}(lambda_4 - lambda_2^2)^{frac{1}{2}}t|u|^{-1}) - u} \ sim |u|^{-3}{t^4/4! + omega(lambda_4 - lambda_2^2)^{frac{1}{2}}lambda_2^ {-frac{1}{2}}t^3/3! - zeta(lambda_4 - lambda_2^2)lambda_2 ^{-1}t^2/2}end{align*} where ω and ζ are independent random variables (rv), ω has a standard normal distribution and ζ has the density z exp (-z), z > 0 . Thus, in the neighborhood of a very low maximum the sample functions are fourth degree polynomials with positive t4-term, symmetrically distributed t3-term, and a negatively distributed t2-term but without t-term. B: Irregular case. r(t) = 1 - λ2t2/2 + λ4t4/4! - λ5|t|5/5! + o(t5) as t → 0, where λ5 > 0 . Now ξ(tu-2) - u ∼ |u|-5{λ2λ5(λ4 - λ22)-1 |t|3/3! + (2λ5)1/2 ω(t) - ζ(λ4 - λ22)λ2 -1t2/2} where ω(t) is a non-stationary normal process whose second derivative is a Wiener process, independent of ζ which has the density z exp (-z), z > 0 . The term λ5|t|5/5! "disturbs" the process in such a way that the order of the distance which can be surveyed is reduced from 1/|u| (in Case A) to 1/|u|2. The results are used in Section 5 to examine the distribution of the wave-length and the crest-to-trough wave-height, i.e., the amplitude, discussed by, among others, Cartwright and Longuet-Higgins, 1956 [1]. One hypothesis, sometimes found in the literature, [10], states that the amplitude has a Rayleigh distribution and is independent of the mean level. According to this hypothesis the amplitude is of the order 1/|u| as u → -∞ while the results of this paper show that it is of the order 1/|u|3. (Less)
[1]
William Feller,et al.
An Introduction to Probability Theory and Its Applications
,
1967
.
[2]
Local maxima of stationary processes
,
1966
.
[3]
G. Lindgren.
Extreme values of stationary normal processes
,
1971
.
[4]
harald Cramer,et al.
Stationary And Related Stochastic Processes
,
1967
.
[5]
Feller William,et al.
An Introduction To Probability Theory And Its Applications
,
1950
.
[6]
S. Berman.
Stationary and Related Stochastic Processes
,
1967
.
[7]
Sverker Sjöström.
On random load analysis
,
1961
.
[8]
S. Rice.
Mathematical analysis of random noise
,
1944
.
[9]
Gerald M. White.
Zeros of Gaussian Noise
,
1958
.
[10]
C. R. Rao,et al.
Linear Statistical Inference and its Applications
,
1968
.
[11]
N. L. Johnson,et al.
Linear Statistical Inference and Its Applications
,
1966
.
[12]
M. Longuet-Higgins,et al.
The statistical distribution of the maxima of a random function
,
1956,
Proceedings of the Royal Society of London. Series A. Mathematical and Physical Sciences.
[13]
D. Slepian,et al.
Large Excursions of Gaussian Processes
,
1959
.