經由研究一個輸入序列的轉換模式來關注結構性變化之偵測與處理。 / Time series data are often subject to uncontrolled or unexpected interventions, from which various types of outlying observations or structure changes are produced. In this article, we focus on detecting and treating structure change events in multiple time series by studying transfer function models with one input series. Monte Carlo simulations will be used to study the performance of the proposed procedures.

Consider a time series model which is stationary apart from a single shift in mean. If the time of a level shift is known, the least squares estimator of the magnitude of this level shift is a minimum variance unbiased estimator. If the time is unknown, however, this estimator is biased. Here, we first carry out extensive simulation studies to determine the relationship between the bias and three parameters of our time series model: the true magnitude of the level shift, the true time point and the autocorrelation of adjacent observations. Thereafter, we use two generalized additive models to generalize the simulation results. Finally, we examine to what extent the bias can be reduced by multiplying the least squares estimator with a shrinkage factor. Our results showed that the bias of the estimated magnitude of the level shift can be reduced when the level shift does not occur close to the beginning or end of the time series. However, it was not possible to simultaneously reduce the bias for all possible time points and magnitudes of the level shift.

The dissertation consists of three chapters on econometric methods related to parameter instability, forecasting and long memory. The first chapter introduces a new frequentist-based approach to forecast time series in the presence of in and out-of-sample breaks in the parameters. We model the parameters as random level shift (RLS) processes and introduce two features to make the changes in parameters forecastable. The first models the probability of shifts according to some covariates. The second incorporates a built-in mean reversion mechanism to the time path of the parameters. Our model can be cast into a non-linear non-Gaussian state-space framework. We use particle filtering and Monte Carlo expectation maximization algorithms to construct the estimates. We compare the forecasting performance with several alternative methods for different series. In all cases, our method allows substantial gains in forecasting accuracy. The second chapter extends the RLS model of Lu and Perron (2010) for the volatility of asset prices. The extensions are in two directions: a) we specify a time-varying probability of shifts as a function of large negative lagged returns; b) we incorporate a mean reverting mechanism so that the sign and magnitude of the jump component change according to the deviations of past jumps from their long run mean. We estimate the model using daily data on four major stock market indices. Compared to competing models, the modified RLS model yields the smallest mean square forecast errors overall. The third chapter proposes a method of inference about the mean or slope of a time trend that is robust to the unknown order of fractional integration of the errors. Our tests have the standard asymptotic normal distribution irrespective of the value of the long-memory parameter. Our procedure is based on using quasi-differences of the data and regressors based on a consistent estimate of the long-memory parameter obtained from the residuals of a least-squares regression. We use the exact local-Whittle estimator proposed by Shimotsu (2010). Simulation results show that our procedure delivers tests with good finite sample size and power, including cases with strong short-term correlations.

<p>Ab-initio results of the core-level shift and the distribution about the average for the 3<i>d</i>_5/2 electrons of Ag, Pd and 2<i>p</i>_3/2 of Cu are presented for the face-centered-cubic AgPd and CuPd random alloys. The complete screening model, which includes both initial and final states effects in the same scheme, has been used in the investigations. </p><p>The alloys have been modeled with a supercell containing 256 atoms. Density-functional theory calculations are carried out using the locally self consistent Green's function approach. </p><p>Results from the calculations clearly shows that the core-level shift distributions characteristic is Gaussian, but the components reveals a substantial difference in the FWHM (Full-Width at Half-Maximum). Comparison between the experimental and the calculated broadening shows a remarkable agreement.</p>

Ab-initio results of the core-level shift and the distribution about the average for the 3d5/2 electrons of Ag, Pd and 2p3/2 of Cu are presented for the face-centered-cubic AgPd and CuPd random alloys. The complete screening model, which includes both initial and final states effects in the same scheme, has been used in the investigations. The alloys have been modeled with a supercell containing 256 atoms. Density-functional theory calculations are carried out using the locally self consistent Green's function approach. Results from the calculations clearly shows that the core-level shift distributions characteristic is Gaussian, but the components reveals a substantial difference in the FWHM (Full-Width at Half-Maximum). Comparison between the experimental and the calculated broadening shows a remarkable agreement.

<p>Theoretical investigations of spectroscopic and magnetic properties of metallic systems in the bulk, as well as in nanostructured materials, have been performed within the density functional theory. The major part of the present work studies the differences between binding energies of electrons tightly bound to the atoms, the so-called core electrons (in contrast with the valence electrons), that is, core-level binding energy shift (CLS). </p><p>By comparison between corresponding elemental core-levels for atoms situated in different chemical environments we obtain fundamental understanding of bonding properties of materials. The method of choice was the complete screening picture, which includes initial and final state effects on the same footing. The usefulness of CLS stems from that it is sensitive to differences in the chemical environment of an atom, which can be affected on one hand by the global composition of e.g. disordered materials, surfaces and interfaces, and on the other hand by the very local environment around an atom. Here CLSs have been obtained for both components in the fcc random alloys AgPd, CuPd, CuNi, CuPt, CuAu, PdAu, NiPd and NiPt. Moreover the model was extended to the Auger kinetic energy shift for the LMM Auger transition in AgPd alloys. Studies were also applied to the near surface and interface regions of PdMn nano structures on Pd(100), thin CuPd and AgPd films on inert Ru(0001), and at interfaces. The disorder broadening on CLS due to local environment effects was calculated in selected alloys.</p><p>A part of the thesis concern investigations related to the magnetic ordering in Invar alloys, including the influence of local environment effects. A study was made for the dependence of effective exchange parameter on the electron concentration, volume and local chemical composition.</p>

Theoretical investigations of spectroscopic and magnetic properties of metallic systems in the bulk, as well as in nanostructured materials, have been performed within the density functional theory. The major part of the present work studies the differences between binding energies of electrons tightly bound to the atoms, the so-called core electrons (in contrast with the valence electrons), that is, core-level binding energy shift (CLS). By comparison between corresponding elemental core-levels for atoms situated in different chemical environments we obtain fundamental understanding of bonding properties of materials. The method of choice was the complete screening picture, which includes initial and final state effects on the same footing. The usefulness of CLS stems from that it is sensitive to differences in the chemical environment of an atom, which can be affected on one hand by the global composition of e.g. disordered materials, surfaces and interfaces, and on the other hand by the very local environment around an atom. Here CLSs have been obtained for both components in the fcc random alloys AgPd, CuPd, CuNi, CuPt, CuAu, PdAu, NiPd and NiPt. Moreover the model was extended to the Auger kinetic energy shift for the LMM Auger transition in AgPd alloys. Studies were also applied to the near surface and interface regions of PdMn nano structures on Pd(100), thin CuPd and AgPd films on inert Ru(0001), and at interfaces. The disorder broadening on CLS due to local environment effects was calculated in selected alloys. A part of the thesis concern investigations related to the magnetic ordering in Invar alloys, including the influence of local environment effects. A study was made for the dependence of effective exchange parameter on the electron concentration, volume and local chemical composition.