• Journal of the Korean Data and Information Science Society
• /
• v.18 no.1
• /
• pp.211-218
• /
• 2007

Nonlinear ARMA model $X_n\;=\;h(X_{n-1},{\cdots},X_{n-p},e_{n-1},{\cdots},e_{n-p})+e_n$ is considered and easy-to-check sufficient condition for strict stationarity of {$X_n$} without some irreducibility or continuity assumption is given. Threshold ARMA(p, q) and momentum threshold ARMA(p, q) models are examined as special cases.

• 제어로봇시스템학회:학술대회논문집
• /
• 1989.10a
• /
• pp.427-430
• /
• 1989

For on-line ARMA spectral estimation, the fast transversal filter algorithm of extended least squares method(ETS FTF) is presented. The projection operator, a key tool for geometric approach, is used in the derivation of the algorithm. ELS FTF is a fast time update recursion which is based on the fact that the correlation matrix of ARMA model satisfies the shift invariance property in each block, and thus it takes 10N+31 MADPR.

• 제어로봇시스템학회:학술대회논문집
• /
• 1992.10a
• /
• pp.923-926
• /
• 1992

ARMA fast Transversal filter(FTF) algorithm solves the extended least squres estimation problems in a very efficient way. But unfortunately, it exhibits a very unstable behavior, due to the accumulation of round-off errors. So, in this paper, two effective method to stabilize ARMA FTF algorithm is proposed. They are based on the analysis of the propagation of the numerical errors according to a first order linear model. The proposed methods modify the numerical properties of the variables responsible for the numerical instability, while proeserving the theoretical form of the algorithm. The proposed algorithms still have the nice complexity properties of the original algorithm, but have a much more stable brhavior.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.15 no.26
• /
• pp.181-187
• /
• 1992

본 논문은 CIMS에서 적응되는 ARMA 공정제어의 새로운 3단계절차를 제안한다. 첫번째 단계는 다변량 ARMA모델을 식별하여 모수를 추정하고, white noise로 진단된 잔차 series에 대하여 다변량 제어통계량(즉, 다변량 Hotelling T$^2$통계량, 다변량 CUSUM, 다변량 EWHA 통계량, 다변량 MA 통계량)등을 계산한다. 마지막으로 본 논문에서 제안한 8가지 다변량 제어통계량을 상호비교하여 이상점을 발견한다.

• Journal of the Korean Society for Aeronautical & Space Sciences
• /
• v.34 no.4
• /
• pp.47-52
• /
• 2006

Using the equivalent ARMA model representation of the mixed random errors, we propose Klaman filter design methods for aided INS(Inertial Navigation System) which contains the gyroscope mixed random errors. At first step, considering the characteristic of indirect feedback Kalman filter used in the aided INS, we perform the time difference of equivalent ARMA model. Next, according to the order of the time differenced ARMA model, we achieve the state space conversion of that by two methods. If the order of AR part is greater than MA part, we use controllable or observable canonical form. Otherwise, we establish the state apace equation via the method that several step ahead predicts are included in the state variable, where we can derive high and low order models depending on the variable which is compensated from gyroscope output. At final step, we include the state space equation of gyroscope mixed random errors into aided INS Kalman filter model. Through the simulation, we show that both the high and low order filter models proposed give less navigation errors compared to the conventional filter which assume the mixed random errors as white noise.

• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.13 no.4
• /
• pp.405-415
• /
• 2000

Time series of wind-induced pressure on a structure are modeled using autoregressive moving average (ARMA) model. In an AR process, the current value of the time series is expressed in terms of a finite, linear combination of the previous values and a white noise. In a MA process, the value of the time series is linearly dependent on a finite number of the previous white noises. The ARMA process is a combination of the AR and MA processes. In this paper, the ARMA models with several different combinations of the AR and MA orders are fitted to the wind-induced pressure time series, and the procedure to select the most appropriate ARMA model to represent the data is described. The maximum likelihood method is used to estimate the model parameters, and the AICC model selection criterion is employed in the optimization of the model order, which is assumed to be a measure of the temporal complexity of the pressure time series. The goodness of fit of the model is examined using the LBP test. It is shown that AR processes adequately fit wind pressure time series.

• Communications for Statistical Applications and Methods
• /
• v.20 no.4
• /
• pp.339-345
• /
• 2013

In this paper, we give a tractable sufficient condition for functional central limit theorem to hold in Markov switching ARMA (p, q) model.

• Proceedings of the Korean Operations and Management Science Society Conference
• /
• 1991.10a
• /
• pp.303-304
• /
• 1991

• Journal of the Korean Data and Information Science Society
• /
• v.20 no.2
• /
• pp.453-458
• /
• 2009

In this paper, we consider Jarque-Bera (JB) normality test for the innovations of ARMA-GARCH models. In financial applications, JB test based on the residuals are routinely used for the normality of ARMA-GARCH innovations without a justification. However, the validity of JB test should be justified in advance of the actual practice (Lee et al., 2009). Through the simulation study, it is found that the validity of JB test depends on the shape of test statistic. Specifically, when the constant term is involved in ARMA model, a certain type of residual based JB test produces severe size distortions.

• 제어로봇시스템학회:학술대회논문집
• /
• 2005.06a
• /
• pp.262-264
• /
• 2005

We propose a filter design method to suppress the effect of gyroscope mixed random errors at INS system level. It is based on the result that mixed random errors can be represented by a single equivalent ARMA model. At first step, the time difference of equivalent ARMA process is performed, which consider the characteristic of indirect feedback Kalman filter used in INS filter. Next, a state space conversion of time differenced ARMA model is achieved. If the order of AR is greater than that of MA, the controllable or observable canonical form is used. Otherwise, we introduce the state equation of which the state variable is composed of the ARMA model output and several step ahead predicts of that. At final step, a complete form state equation is presented. The simulation results shows that the proposed method gives less transient error and better convergence compared to the conventional filter which assume the mixed random errors as white noise.