중앙대학교 응용통계학과 데이터과학연구소에서 U.S. Bureau of Labor Statistics에 계신 Peter Zadrozny 박사님을 모시고 아래와 같이 세미나를 개최합니다.
많은 참석 부탁드립니다.
많은 참석 부탁드립니다.
세미나 참석하시는 분들에게 식사와 주차권이 발급되지 않습니다.
이 점 양해 부탁드립니다.
1. 연 사:
Peter Zadrozny
Research Economist
U.S. Bureau of Labor Statistics
Division of Price and Index Number Research
2. 일 시: 2015년 6월 24일 수요일 오후 4시
3. 장 소: 중앙대학교 법학관 (303동) 702호
4. 제 목: Extended Yule-Walker Identification of VARMA Models with Single- or Mixed-Frequency Data
5. 초 록:
Chen and Zadrozny (1998) developed the linear extended Yule-Walker (XYW) method for determining the parameters of a vector autoregressive (VAR) model using available covariances of mixed-frequency data (MFD). XYW takes data covariances as inputs and determines AR parameters as outputs. If the covariance inputs are true population values and the outputs are unique, then, the outputs are true parameter values, i.e., the model is identified; if the covariance inputs are consistent sample estimates and the outputs are unique, then, the outputs are consistent parameter estimates. The present paper extends XYW to "extended XYW" (X^2YW) for determining all ARMA parameters of a VARMA model using available covariances of single-frequency data (SFD) or MFD. The paper also proves, under stated conditions on parameters, that the outputs are unique, so that the VARMA model is identified for true population-covariance inputs and is consistently estimated for consistent sample-covariance inputs. X^2YW solves one linear equation to determine the AR parameters and solves two linear equations and does one spectral factorization to determine the MA parameters.
Chen and Zadrozny (1998) developed the linear extended Yule-Walker (XYW) method for determining the parameters of a vector autoregressive (VAR) model using available covariances of mixed-frequency data (MFD). XYW takes data covariances as inputs and determines AR parameters as outputs. If the covariance inputs are true population values and the outputs are unique, then, the outputs are true parameter values, i.e., the model is identified; if the covariance inputs are consistent sample estimates and the outputs are unique, then, the outputs are consistent parameter estimates. The present paper extends XYW to "extended XYW" (X^2YW) for determining all ARMA parameters of a VARMA model using available covariances of single-frequency data (SFD) or MFD. The paper also proves, under stated conditions on parameters, that the outputs are unique, so that the VARMA model is identified for true population-covariance inputs and is consistently estimated for consistent sample-covariance inputs. X^2YW solves one linear equation to determine the AR parameters and solves two linear equations and does one spectral factorization to determine the MA parameters.