Browsing by Author "Chou, PH"

Now showing 1 - 5 of 5

Detecting mutual fund timing ability using the threshold model
(2005-10-20)
Chou, PH
;
Chung, HM
;
Sun, EY
;
管理科學系
;
資訊管理與財務金融系註：原資管所+財金所
This paper proposes a new method based on threshold regression to test mutual fund market-timing abilities. The traditional Henriksson and Merton model is shown to represent only a special case within the proposed model. The potential bias of using the traditional model is demonstrated and it is argued that the proposed model provides more accurate inferences on the market-timing effects or mutual funds. The empirical results for a set of randomly-selected US mutual funds indicate the superior performance of the proposed method in detecting the market-timing ability.
Efficient polygonization of CSG solids using boundary tracking
(1997-11-01)
Chung, CW
;
Chuang, JH
;
Chou, PH
;
資訊工程學系
;
Department of Computer Science
In this paper, we propose a procedure that directly polygonizes the boundary surface of CSG solids. The procedure consists of a preprocessing step and a polygonization process followed by a postprocessing that recovers rounded edges and corners. In the preprocessing step, the minimum bounding volumes, called S-bounds, of all nodes in the given CSG tree are computed and then used as a basis for subdividing the S-bound of the tree's root. In addition, a regular grid embedded on the root's S-bound is constructed. The polygonization is performed in such a way that among leaf voxels of the space subdivision only voxels that overlap the solid's boundary are traversed and inside such voxels only grid cells transversal to the solid's boundary are tracked. The surface-edge intersections and vertex normals are computed directly from the exact boundary surfaces of the CSG solid. In the postprocessing step, rounded edges and corners are detected and recovered using primitive geometry. (C) 1997 Elsevier Science Ltd. All rights reserved.
Efficient polygonization of CSG solids using boundary tracking
(1997-11-01)
Chung, CW
;
Chuang, JH
;
Chou, PH
;
資訊工程學系
;
Department of Computer Science
In this paper, we propose a procedure that directly polygonizes the boundary surface of CSG solids. The procedure consists of a preprocessing step and a polygonization process followed by a postprocessing that recovers rounded edges and corners. In the preprocessing step, the minimum bounding volumes, called S-bounds, of all nodes in the given CSG tree are computed and then used as a basis for subdividing the S-bound of the tree's root. In addition, a regular grid embedded on the root's S-bound is constructed. The polygonization is performed in such a way that among leaf voxels of the space subdivision only voxels that overlap the solid's boundary are traversed and inside such voxels only grid cells transversal to the solid's boundary are tracked. The surface-edge intersections and vertex normals are computed directly from the exact boundary surfaces of the CSG solid. In the postprocessing step, rounded edges and corners are detected and recovered using primitive geometry. (C) 1997 Elsevier Science Ltd. All rights reserved.
Isomers of OCS2: IR absorption spectra of OSCS and O(CS2) in solid ar
(2004-12-22)
Lo, WJ
;
Chen, HF
;
Chou, PH
;
Lee, YP
;
應用化學系
Irradiation of an Ar matrix sample containing O-3 and CS2 with a KrF excimer laser at 248 nm yielded new lines at 1402.1 (1404.7), 1056.2 (1052.7), and 622.3 (620.5) cm-1; numbers in parentheses correspond to species in a minor matrix site. Secondary photolysis at 308 nm diminished these lines and produced mainly OCS and SO2. Annealing of this matrix to 30 K yielded a second set of new lines at 1824.7 and 617.8 cm-1. The first set of lines are assigned to C=S stretching, O-S stretching, and S-C stretching modes of carbon disulfide S-oxide, OSCS; and the second set of lines are assigned to C=O stretching and OCS bending modes of dithiiranone, O(CS2), respectively, based on results of S-34- and O-18-isotopic experiments and quantum-chemical calculations. These calculations using density-functional theory (B3LYP/aug-cc-pVTZ) predict four stable isomers of OCS2: O(CS2), SSCO, OSCS, and SOCS, listed in order of increasing energy. According to calculations, O(CS2) has a cyclic CS2 moiety and is the most stable isomer of OCS2. OSCS is planar, with bond angles angleOSCcongruent to111.9degrees and angleSCScongruent to177.3degrees; it is less stable than SSCO and O(CS2) by similar to102 and 154 kJ mol-1, respectively, and more stable than SOCS by similar to26 kJ mol-1. Calculated vibrational wave numbers, IR intensities, S-34- and O-18-isotopic shifts for OSCS and O(CS2) fit satisfactorily with experimental results. (C) 2004 American Institute of Physics.
Nanoprobe-based affinity mass spectrometry for selected protein profiling in human plasma
(2005-09-15)
Chou, PH
;
Chen, SH
;
Liao, HK
;
Lin, PC
;
Her, GR
In recent decades, magnetic nanoparticles have emerged as a promising new platform in biomedical applications, particularly bioseparations. We have developed an immunoassay using antibody-conjugated magnetic nanoparticles as an efficient affinity probe to simultaneously preconcentrate and isolate targeted antigens from biological media. We combined this probe with matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI MS) to profile proteins in diluted human plasma. The nanoparticles were designed to detect several disease-associated proteins and could be used directly in MALDI MS without an elution step, thereby facilitating multiple antigen screening and the characterization of antigen variants. Plasma antigens bound rapidly (similar to 10 min) to the antibody-conjugated nanoparticles, allowing the assay to be performed within 20 min. With sensitivity of detection in the femtomole range, the nanoscale immunoassay is superior to assays using microscale particles. We applied our method to comparative protein profiling of patients with gastric cancer and healthy individuals and found differential protein expression levels associated with the disease as well as individuals. Given the flexibility of manipulating functional groups on the nanoprobes, their low cost, robustness, and simplicity of the assay, our approach shows promise for targeted proteome profiling in clinical settings.