Browsing by Author "Cheng, Ching-Wei"

Now showing 1 - 3 of 3

DOUBLY CONSTRAINED FACTOR MODELS WITH APPLICATIONS
(2016-10)
Tsai, Henghsiu
;
Tsay, Ruey S.
;
Lin, Edward M. H.
;
Cheng, Ching-Wei
;
資訊管理與財務金融系註：原資管所+財金所
This paper focuses on factor analysis of multivariate time series. We propose statistical methods that enable analysts to leverage their prior knowledge or substantive information to sharpen the estimation of common factors. Specifically, we consider a doubly constrained factor model that enables analysts to specify both row and column constraints of the data matrix to improve the estimation of common factors. The row constraints may represent classifications of individual subjects whereas the column constraints may show the categories of variables. We derive both the maximum likelihood and least squares estimates of the proposed doubly constrained factor model and use simulations to study the performance of the analysis in finite samples. The Akaike information criterion is used for model selection. Monthly U.S. housing start data from nine geographical divisions are used to demonstrate the application of the proposed model.
DOUBLY CONSTRAINED FACTOR MODELS WITH APPLICATIONS
(2016-10-01)
Tsai, Henghsiu
;
Tsay, Ruey S.
;
Lin, Edward M. H.
;
Cheng, Ching-Wei
;
資訊管理與財務金融系註：原資管所+財金所
This paper focuses on factor analysis of multivariate time series. We propose statistical methods that enable analysts to leverage their prior knowledge or substantive information to sharpen the estimation of common factors. Specifically, we consider a doubly constrained factor model that enables analysts to specify both row and column constraints of the data matrix to improve the estimation of common factors. The row constraints may represent classifications of individual subjects whereas the column constraints may show the categories of variables. We derive both the maximum likelihood and least squares estimates of the proposed doubly constrained factor model and use simulations to study the performance of the analysis in finite samples. The Akaike information criterion is used for model selection. Monthly U.S. housing start data from nine geographical divisions are used to demonstrate the application of the proposed model.
Trap-state density in continuous-wave laser-crystallized single-grainlike silicon transistors
(2006-06-05)
Lin, Yu-Ting
;
Chen, Chih
;
Shieh, Jia-Min
;
Lee, Yao-Jen
;
Pan, Ci-Ling
This investigation characterizes electrical characteristics of continuous-wave green laser-annealed single-grainlike silicon thin-film transistors in relation to trap-state densities. As laser power increases, highly crystalline channels form, reducing tail-state densities to as low as 3 X 10(19) eV-' cm(-3). This occurrence is responsible for high field-effect electron mobility of 284 cm(2)/Vs. In contrast, increasing laser power initially reduces the deep-state density and then increases it to 3 X 1016 eV(-1) cm(-3). This reversal in deep-state density and thus in the subthreshold slope as well as a saturating reduction in threshold voltage are associated with the formation of extra interface defects caused by laser-crystallization-enhanced surface roughness. (c) 2006 American Institute of Physics.