Author: Ulrich K. Müller Publisher: ISBN: Category : Languages : en Pages : 0
Book Description
The paper investigates inference in nonlinear and non-Gaussian models with moderately time varying parameters. We show that for many decision problems, the sample information about the parameter path can be summarized by an artificial linear and Gaussian model, at least asymptotically. The approximation allows for computationally convenient path estimators and parameter stability tests. Also, in contrast to standard Bayesian techniques, the artificial model can be robustified so that in misspecified models, decisions about the path of the (pseudo-true) parameter remain as good as in a corresponding correctly specified model.
Author: A Clifford Cohen Publisher: CRC Press ISBN: 1000147231 Category : Mathematics Languages : en Pages : 312
Book Description
Offers an applications-oriented treatment of parameter estimation from both complete and censored samples; contains notations, simplified formats for estimates, graphical techniques, and numerous tables and charts allowing users to calculate estimates and analyze sample data quickly and easily. Anno
Author: James Thomas McClave Publisher: ISBN: Category : Time-series analysis Languages : en Pages : 206
Book Description
Many techniques are available for estimating the parameters of linear stationary time series. The effect of some of these estimators on the least squares predictor for some future value of the series is examined. We have obtained approximations for the increase in prediction error due to parameter estimation in several cases for which no exact expression could be found. An efficient estimator for the parameters in a first order autoregressive-moving average model is developed by making use of a linear function of the autocorrelations. For more general models we conclude that efficient estimation is so difficult to attain that first consideration in many estimation prediction problems should be given to ease of calculation. Numerous unsolved estimation and prediction problems remain for non-stationary time series. We consider the logistic growth process, which is used extensively as an economic and population growth model, in detail. Current estimation procedures for the logistic process' parameters make no reference to an error structure. We propose a probability structure consistent with the realistic properties of the series. We then use this structure to obtain estimators from three different observational standpoints: (1) that of observation at equidistant time points, (2) that of continuous observation, and (3) that of arrival time observation. For observation types (1) and (2) we have used a modification of maximum likelihood procedures to obtain estimators having most of the usual properties associated with maximum likelihood estimators. A computer program was written for type (1) to solve the intractable estimation equations, using the Newton-Rhapson procedure. Observation of arrival times is shown not to lead to any useful estimation procedures. We also examine the effect of the estimators calculated from the observations taken at equal time intervals (type (1) above) on the error of prediction. The procedure developed for observation type (1) is then compared by means of example to an estimation procedure developed by Rhodes. The logistic model is also fitted by each method to the population of conterminous United States. The estimates are then used to obtain predictions of the population of conterminous United States in 1970. We conclude from the results that the effort required to calculate the maximum likelihood estimates is worthwhile. We further conjecture that the methods developed may be applicable to other growth processes.
Author: K. Dzhaparidze Publisher: Springer Science & Business Media ISBN: 1461248426 Category : Mathematics Languages : en Pages : 331
Book Description
. . ) (under the assumption that the spectral density exists). For this reason, a vast amount of periodical and monographic literature is devoted to the nonparametric statistical problem of estimating the function tJ( T) and especially that of leA) (see, for example, the books [4,21,22,26,56,77,137,139,140,]). However, the empirical value t;; of the spectral density I obtained by applying a certain statistical procedure to the observed values of the variables Xl' . . . , X , usually depends in n a complicated manner on the cyclic frequency). . This fact often presents difficulties in applying the obtained estimate t;; of the function I to the solution of specific problems rela ted to the process X . Theref ore, in practice, the t obtained values of the estimator t;; (or an estimator of the covariance function tJ~( T» are almost always "smoothed," i. e. , are approximated by values of a certain sufficiently simple function 1 = 1
Author: Christos Floros Publisher: Springer ISBN: 3319591029 Category : Business & Economics Languages : en Pages : 323
Book Description
This book sheds new light on the Greek economic challenges and helps readers understand the current debt crisis. Chapters from leading experts in the field identify and outline potential solutions to the on-going decline of the Greek economy by considering both Eurozone-adopted current policy framework explanations and potential alternative explanations. In contrast to the standard chronological approach toward the Greek debt crisis typically adopted by other texts, this book draws on the experience and views of specialized economists and offers divergent opinions that could potentially form alternative solutions. It will be of interest to researchers and academics interested in the Greek economy, modern financial modelling, and risk management.
Author: Peter J. Brockwell Publisher: Walter de Gruyter GmbH & Co KG ISBN: 3111325032 Category : Mathematics Languages : en Pages : 522
Book Description
This book provides a self-contained account of continuous-parameter time series, starting with second-order models. Integration with respect to orthogonal increment processes, spectral theory and linear prediction are treated in detail. Lévy-driven models are incorporated, extending coverage to allow for infinite variance, a variety of marginal distributions and sample paths having jumps. The necessary theory of Lévy processes and integration of deterministic functions with respect to these processes is developed at length. Special emphasis is given to the analysis of continuous-time ARMA processes.
Author: Ulrich K. Müller
Author: Shuwen Zhao
Author: A Clifford Cohen
Author: James Thomas McClave
Author: K. Dzhaparidze
Author: Margaret Elizabeth Andrews
Author: Douglas Gardiner Steigerwald
Author: Christos Floros
Author: Peter J. Brockwell
Author: Fumio Hayashi