• 제목/요약/키워드: Covariance stationary

검색결과 28건 처리시간 0.022초

A Note on the Dependence Conditions for Stationary Normal Sequences

  • Choi, Hyemi
    • Communications for Statistical Applications and Methods
    • /
    • 제22권6호
    • /
    • pp.647-653
    • /
    • 2015
  • Extreme value theory concerns the distributional properties of the maximum of a random sample; subsequently, it has been significantly extended to stationary random sequences satisfying weak dependence restrictions. We focus on distributional mixing condition $D(u_n)$ and the Berman condition based on covariance among weak dependence restrictions. The former is assumed for general stationary sequences and the latter for stationary normal processes; however, both imply the same distributional limit of the maximum of the normal process. In this paper $D(u_n)$ condition is shown weaker than Berman's covariance condition. Examples are given where the Berman condition is satisfied but the distributional mixing is not.

비정체형 2차원 다공성 매질의 대수투수계수-수두 교차공분산에 관한 연구 (A Study on Logconductivity-Head Cross Covariance in Two-Dimensional Nonstationary Porous Formations)

  • 성관제
    • 물과 미래
    • /
    • 제29권5호
    • /
    • pp.215-222
    • /
    • 1996
  • 본 논문에서는 다공성 매질의 특수율이 비정체형인 경우 대수투수계수-수두 교차공분산에 관한 식을 유도하였으며, 이 교차공분산은 수두분포로부터 특수장의 통계학적 특성을 유추하는데(inverse problem) 매우 중요한 역할을 담당한다. 비정체형 대수투수계수는 일정한 선형경향과 정체형인 미소 변동의 합으로 구성되었으며, 2차원 포화대수층에서 정상 유동문제를 추계학적으로 해석하여 수두분포를 얻었고 이로부터 교차공분산을 유도하였다. 투수계수의 상관함수가 가우스분포를 가지고 그 경향이 수두 경사와 평행이거나 직교하는 두 가지 경우에 대하여 교차공분산을 살펴 본 결과, 투수장의 경향이 주 흐름방향과 평행한 경우 흐름방향 쪽만 제외하고는 정체형임이 밝혀졌다. 또한, 흐름방향과 직교하는 쪽으로의 교차공분산은 정체형 모델 결과와 달리 영이 아님를 알 수 있었다. 따라서 지하수 유동이나 오염물질 확산문제를 다룰 경우, 투수계수장에 어떤 경향이 존재한다고 의심될 때에는 반드시 그 경향을 해석과정에 포함시켜야 한다.

  • PDF

A Note on the Covariance Matrix of Order Statistics of Standard normal Observations

  • Lee, Hak-Myung
    • Communications for Statistical Applications and Methods
    • /
    • 제7권1호
    • /
    • pp.285-290
    • /
    • 2000
  • We noted a property of a stationary distribution on the matrix C, which is the covariance matrix of order statistics of standard normal distribution That is the sup norm of th powers of C is ee' divided by its dimension. The matrix C can be taken as a transition probability matrix in an acyclic Markov chain.

  • PDF

ON STATIONARY GAUSSIAN SECOND ORDER MARKOV PROCESSES

  • Park, W.J.;Hsu, Y.S.
    • Kyungpook Mathematical Journal
    • /
    • 제19권2호
    • /
    • pp.249-255
    • /
    • 1979
  • In this paper we give a characterization of Stationary Gaussian 2nd order Markov processes in terms of its covariance function $R({\tau})=E[X(t)X(t+{\tau})]$ and also give some relationship among quasi-Markov, Markov and 2nd order Markov processes.

  • PDF

유체-구조물 상호작용을 고려한 면진구조물의 추계학적 응답해석 (Stochastic Analysis of Base-Isolated Pool Structure Considering Fluid-Structure Interaction Effects)

  • 고현무;김재관;박관순;하동호
    • 대한토목학회논문집
    • /
    • 제14권3호
    • /
    • pp.463-472
    • /
    • 1994
  • Random 지반운동에 대한 면진수조구조물 응답의 추계학적 해석방법을 연구하였다. 유연한 벽체와 내부유체간의 유체구조물 상호작용은 유체운동의 유한요소 모델링에 의하여 얻어지는 부가질량행렬 형태로 고려되었다. 정상(定常)(Stationary)지반운동으로는 수정된 Clough-Penzien Spectral Model이 사용되었으며, 비정상(非定常)(Nonstationary)지반운동으로는 상기모델에 포락함수를 적용한 모델을 사용하였다. 운동을 지배하는 Lyapunov Covariance Matrix 미분방정식의 해를 구하여 두 종류 면진시스템의 정상응답 및 비정상응답을 해석하였다.

  • PDF

A Note on Stationary Linearly Positive Quadrant Dependent Sequences

  • Kim, Tae-Sung
    • Journal of the Korean Statistical Society
    • /
    • 제24권1호
    • /
    • pp.249-256
    • /
    • 1995
  • In this note we prove an invariance principle for strictly stationary linear positive quadrant dependent sequences, satifying some assumption on the covariance structure, $0 < \sum Cov(X_1,X_j) < \infty$. This result is an extension of Burton, Dabrowski and Dehlings' invariance principle for weakly associated sequences to LPQD sequences as well as an improvement of Newman's central limit theorem for LPQD sequences.

  • PDF

유연성을 가지는 비행체를 위한 속도/방위각 정합 전달 정렬 알고리즘 설계 (Design of Transfer Alignment Algorithm with Velocity and Azimuth Matching for the Aircraft Having Wing Flexibility)

  • 강석태
    • 한국군사과학기술학회지
    • /
    • 제26권3호
    • /
    • pp.214-226
    • /
    • 2023
  • A transfer alignment is used to initialize, align, and calibrate a SINS(Slave INS) using a MINS(Master INS) in motion. This paper presents an airborne transfer alignment with velocity and azimuth matching to estimate inertial sensor biases under the wing flexure influence. This study also considers the lever arm, time delay and relative orientation between MINS and SINS. The traditional transfer alignment only uses velocity matching. In contrast, this paper utilizes the azimuth matching to prevent divergence of the azimuth when the aircraft is stationary or quasi-stationary since the azimuth is less affected by the wing flexibility. The performance of the proposed Kalman filter is analyzed using two factors; one is the estimation performance of gyroscope and accelerometer bias and the other is comparing aircraft dynamics and attitude covariance. The performance of the proposed filter is verified using a long term flight test. The test results show that the proposed scheme can be effectively applied to various platforms that require airborne transfer alignment.

Some Notes on the Fourier Series of an Almost Periodic Weakly Stationary Process

  • You, Hi-Se
    • Journal of the Korean Statistical Society
    • /
    • 제3권1호
    • /
    • pp.13-16
    • /
    • 1974
  • In my former paper [3] I defined an almost periodicity of weakly sationary random processes (a.p.w.s.p.) and presented some basic results of it. In this paper I shall present some notes on the Fourier series of an a.p.w.s.p., resulting from [3]. All the conditions at the introduction of [3] are assumed to hold without repreating them here. The essential facts are as follows : The weakly stationary process $X(t,\omega), t\in(-\infty,\infty), \omega\in\Omega$, defined on a probability space $(\Omega,a,P)$, has a spectral representation $$X(t,\omega)=\int_{-\infty}^{infty}{e^{it\lambda\xi}(d\lambda,\omega)},$$ where $\xi(\lambda)$ is a random measure. Then, the continuous covariance $\rho(\mu) = E(X(t+u) X(t))$ has the form $$\rho(u)=\int_{-\infty}^{infty}{e^{iu\lambda}F(d\lambda)},$$ $E$\mid$\xi(\lambda+0)-\xi(\lambda-0)$\mid$^2 = F(\lambda+0) - F(\lambda-0) \lambda\in(-\infty,\infty)$, assumimg that $\rho(u)$ is a uniformly almost periodic function.

  • PDF