Matching Items (4)
Description
Estimating cointegrating relationships requires specific techniques. Canonical correlations are used to determine the rank and space of the cointegrating matrix. The vectors used to transform the data into canonical variables have an eigenvector representation, and the associated canonical correlations have an eigenvalue representation. The number of cointegrating relations is chosen based upon a theoretical difference in the convergence rates of the eignevalues. The number of cointegrating relations is consistently estimated using a threshold function which places a lower bound on the eigenvalues associated with cointegrating relations and an upper bound on the eigenvalues on the eigenvalues not associated with cointegrating relations. The proposed estimator performs better with a large number of cross-sectional observations and moderate time series length.
ContributorsNowak, Adam (Author) / Ahn, Seung C (Thesis advisor) / Liu, Crocker (Committee member) / Kallberg, Jarl (Committee member) / Arizona State University (Publisher)
Created2012
Description
The computation of the fundamental mode in structural moment frames provides valuable insight into the physical response of the frame to dynamic or time-varying loads. In standard practice, it is not necessary to solve for all n mode shapes in a structural system; it is therefore practical to limit the system to some determined number of r significant mode shapes. Current building codes, such as the American Society of Civil Engineers (ASCE), require certain class of structures to obtain 90% effective mass participation as a way to estimate the accuracy of a solution for base shear motion. A parametric study was performed from the collected data obtained by the analysis of a large number of framed structures. The purpose of this study was the development of rules for the required number of r significant modes to meet the ASCE code requirements. The study was based on the implementation of an algorithm and a computer program developed in the past. The algorithm is based on Householders Transformations, QR Factorization, and Inverse Iteration and it extracts a requested s (s<< n) number of predominate mode shapes and periods. Only the first r (r < s) of these modes are accurate. To verify the accuracy of the algorithm a variety of building frames have been analyzed using the commercially available structural software (RISA 3D) as a benchmark. The salient features of the algorithm are presented briefly in this study.
ContributorsGrantham, Jonathan (Author) / Fafitis, Apostolos (Thesis advisor) / Attard, Thomas (Committee member) / Houston, Sandra (Committee member) / Hjelmstad, Keith (Committee member) / Arizona State University (Publisher)
Created2014
Description
Eigenvalues of the Gram matrix formed from received data frequently appear in sufficient detection statistics for multi-channel detection with Generalized Likelihood Ratio (GLRT) and Bayesian tests. In a frequently presented model for passive radar, in which the null hypothesis is that the channels are independent and contain only complex white Gaussian noise and the alternative hypothesis is that the channels contain a common rank-one signal in the mean, the GLRT statistic is the largest eigenvalue $\lambda_1$ of the Gram matrix formed from data. This Gram matrix has a Wishart distribution. Although exact expressions for the distribution of $\lambda_1$ are known under both hypotheses, numerically calculating values of these distribution functions presents difficulties in cases where the dimension of the data vectors is large. This dissertation presents tractable methods for computing the distribution of $\lambda_1$ under both the null and alternative hypotheses through a technique of expanding known expressions for the distribution of $\lambda_1$ as inner products of orthogonal polynomials. These newly presented expressions for the distribution allow for computation of detection thresholds and receiver operating characteristic curves to arbitrary precision in floating point arithmetic. This represents a significant advancement over the state of the art in a problem that could previously only be addressed by Monte Carlo methods.
ContributorsJones, Scott, Ph.D (Author) / Cochran, Douglas (Thesis advisor) / Berisha, Visar (Committee member) / Bliss, Daniel (Committee member) / Kosut, Oliver (Committee member) / Richmond, Christ (Committee member) / Arizona State University (Publisher)
Created2019
Description
Presented is a design approach and test of a novel compact waveguide that demonstrated the outer dimensions of a rectangular waveguide through the introduction of parallel raised strips, or flanges, which run the length of the rectangular waveguide along the direction of wave propagation. A 10GHz waveguide was created with outer dimensions of a=9.0mm and b=3.6mm compared to a WR-90 rectangular waveguide with outer dimensions of a=22.86mm and b=10.16mm which the area is over 7 times the area. The first operating bandwidth for a hollow waveguide of dimensions a=9.0mm and b=3.6mm starts at 16.6GHz a 40% reduction in cutoff frequency.
The prototyped and tested compact waveguide demonstrated an operating close to the predicted 2GHz with predicted vs measured injection loss generally within 0.25dB and an overall measured injection loss of approximately 4.67dB/m within the operating bandwidth.
The prototyped and tested compact waveguide demonstrated an operating close to the predicted 2GHz with predicted vs measured injection loss generally within 0.25dB and an overall measured injection loss of approximately 4.67dB/m within the operating bandwidth.
ContributorsJones, Jimmy, Ph.D (Author) / Pan, George (Thesis advisor) / Palais, Joseph (Committee member) / Aberle, James T., 1961- (Committee member) / Young, William (Committee member) / Arizona State University (Publisher)
Created2019