Browsing by Subject "Monte Carlo method."

Now showing 1 - 6 of 6

Open Access
Evaluation of the Goldfeld-Quandt test and alternatives
(1994) Tomak, Kerem
In this study, the widely used Coldfeld-C^uandt test for lieterosk('da.sticity in the linear regression model is evaluated. VV(' reduce the dimension of the data spa.ce that is needed lor tin' computaticui of tlu' t('sts. VVe tlu'ii compa.r(‘ the pi'rformaiK'es of tin' Likelihood Ratio and tin* Cloldh'ld-C^uandt tests by using stringency measure. The problem of analytically non-tractable distribution function in the case of the Likelihood Ratio test is overcome by employing Monte Carlo methods. It is observed that the Likelihood Ratio test is better in most of the cases than the Goldfeld-Quandt test.
Open Access
High field transport phenomena in wide bandgap semiconductors
(2003) Sevik, Cem
The Ensemble Monte Carlo (EMC) method is widely used in the field of computational electronics related to the simulation of the state of the art devices. Using this technique our specific intention is to scrutinize the high-field transport phenomena in wide bandgap semiconductors (Such as GaN, AlGaN and AlN). For this purpose, we have developed an EMC-based computer code. After a brief introduction to our methodology, we present detailed analysis of three different types of devices, operating under high-field conditions, namely, unipolar n-type structures, avalanche photodiodes (APD) and finally the Gunn diodes. As a testbed for understanding impact ionization and hot electron effects in sub-micron sized GaN, AlN and their ternary alloys, an n +−n−n + channel device is employed having a 0.1 µm-thick n region. The time evolution of the electron density along the device is seen to display oscillations in the unintentionally doped n-region, until steady state is established. The fermionic degeneracy effects are observed to be operational especially at high fields within the anode n +-region. For AlxGa1−xNbased systems, it can be noted that due to alloy scattering, carriers cannot acquire the velocities attained by the GaN and AlN counterparts. Next, multiplication and temporal response characteristics under a picosecond pulsed optical illumination of p +-n-n + GaN and n-type Schottky Al0.4Ga0.6N APDs are analyzed. For the GaN APD, our simulations can reasonably reproduce the available measured data without any fitting parameters. In the case of AlGaN, the choice of a Schottky contact APD is seen to improve drastically the field confinement resulting in satisfactory gain characteristics. Moreover, alloy scattering is seen to further slow down the temporal response while displacing the gain threshold to higher fields. Finally, the dynamics of large-amplitude Gunn domain oscillations from 120 GHz to 650 GHz are studied in detail by means of extensive EMC simulations. The basic operation is checked under both impressed single-tone sinusoidal bias and external tank circuit conditions. The width of the doping-notch is observed to enhance higher harmonic efficiency at the expense of the fundamental frequency up to a critical value, beyond which sustained Gunn oscillations are ceased. The degeneracy effects due to the Pauli Exclusion principle and the impact ionization are also considered but observed to have negligible effect within the realistic operational bounds. Finally, the effects of lattice temperature, channel doping and DC bias on the RF conversion efficiency are investigated
Open Access
Joint test for structural model specification
(2006) Yüksel, Serkan
Aim of this thesis is to propose a test statistic that can test for true structural model in time series. Main concern of the thesis is to suggest a test statistic, which has joint null of unit root and no structural break (difference stationary model). When joint null hypothesis is rejected, source of deviation from the null model may be structural break or (and) stationarity. Sources of the deviation correspond to different structural models: Pure stationary model, trendbreak stationary model and trend-break with unit root model. The thesis suggests a test statistic that can discriminate null model from alternative models and more importantly, one alternative model from another. The test statistic that is proposed in the thesis is able to detect specific source of deviation from the null model. By doing so, we can determine the true structure model in time series. The thesis compares power properties of the test statistic that is proposed with the most favorable test in the literature. Simulation results indicate the power dominance over the test statistics in the literature. Moreover, we are able to specify true alternative model.
Open Access
Quality measurement plan using Monte Carlo methods
(1997) Zouaoui, Faker
This study considers the Quality Measurement Plan (QMP), a system implemented for reporting the quality assurance audit results to Bell system management. QMP is derived from a new Bayesian approach to the empirical Bayes problem for Poisson observations. It uses both the current and past data to compute estimates for the quality of the current production. The QMP estimator developed by Hoadley in 1981 is based on many complicated approximations. Sampling approaches such as the Gibbs sampler and Importance-sampling are alternative techniques that avoid these approximations and permit the computation of the quality estimates through Monte Carlo methods. Here we discuss the approaches and the algorithms for implementing some Monte Carlo-based approaches on the QMP model. We also show via simulation that although the QMP algorithm can be computationally more convenient, the sampling approaches mentioned above give more accurate estimates of current quality.
Open Access
Variational Monte Carlo calculations for Bose-Hubbard model based on projected wavefunctions
(2014) Koç, Fulya
Bose-Hubbard model is mainly used to describe and study the interactions between neutral atomic gases trapped in an optical lattice [1] and Josephson junction arrays [2]. It is one of the toy models to understand quantum phase transitions, i.e. a phase transition exists between the Mott insulator state and the super- fluid state. Analytical solutions are limited to obtaining the ground state energy for small systems, whereas, computational studies can be done for larger system sizes. We applied the variational Monte Carlo method to the Bose-Hubbard model based on projected wavefunctions, i.e. Baeriswyl and Gutzwiller-Baeriswyl. Even though our method can be applicable to any dimension, we only consider the one dimensional case in this thesis. We expressed observables in forms of averages over configurations to which we can apply Monte Carlo sampling techniques. Our results for both Baeriswyl and Gutzwiller projections are in qualitatively good agreement with the known calculations of the phase diagram [3,4]. Furthermore, we introduced a new method, apart from other known methods [5, 6], based on the Drude weight [7–9] to calculate the superfluid fraction, which can also be extended to observe BCS superconductivity [10].
Open Access
Variational Monte Carlo study of two dimensional charged bosons
(2014-09) Karakuzu, Seher
We studied the ground state properties of 2D charged Bosons interacting via Coulomb potential using the Variational Monte Carlo method. We start with the paper of McMillan [1] in which there are 3D Bosons interacting via Lennard-Jones potential. We calculate ground state energy and pair distribution function of the system and our results are compared with the original work. We also reproduce the paper of Liu et al. [2] and study the 2D Bosons interacting with the same potential. Our results are compared with the original work. In order to evaluate long-range potentials in periodic systems we introduce Ewald summation method for 2D system and Natoli-Ceperley method which evaluates potentials optimally. We compare our Natoli-Ceperley results with the Holzmann et al. [3]. We also investigate 2D charged Boson system interacting via Coulomb potential using the RPA pseudo potential. Our results are compared with de Palo et al. [4]. We try to optimize the wave function further introducing a Kinetic Energy projection to the original RPA wave function. We show that it is very difficult to optimize it further since it is a very good wave function.