Titlebook: Robustness in Statistical Forecasting; Yuriy Kharin Book 2013 Springer International Publishing Switzerland 2013 62-02, 62M20, 62M10, 62G3

interference

Robustness of Time Series Forecasting Based on Regression Models, the .-admissible distortion level. Robust forecasting statistics are constructed by using Huber estimators and a specially chosen type of M-estimators for the regression function parameters. A local-median forecasting algorithm is proposed to mitigate the influence of outliers under regression models, and its robustness is evaluated.

騷擾

Optimality and Robustness of ARIMA Forecasting,lgorithms is evaluated under the following distortion types: parametric model specification errors, functional distortions of the innovation process in the mean value, heteroscedasticity, AO and IO outliers, bilinear autoregression distortions.

不能強(qiáng)迫我

Optimality and Robustness of Vector Autoregression Forecasting Under Missing Values,nd model specification errors. In the case of parametric prior uncertainty, a consistent forecasting statistic and an asymptotic expansion of the corresponding forecast risk are obtained. The chapter is concluded by considering plug-in forecasting under simultaneous influence of outliers and missing values.

消極詞匯

Performance and Robustness Characteristics in Statistical Forecasting,ructed. Robustness of statistical forecasting techniques is defined in terms of the following robustness characteristics: the guaranteed (upper) risk, the risk instability coefficient, the .-admissible distortion level.

reflection

Book 2013lems; .- evaluating the robustness for traditional forecasting procedures under distortions; .- obtaining the maximal distortion levels that allow the “safe” use of the traditional forecasting algorithms; .-?creating new robust forecasting procedures to arrive at risks that are less sensitive to definite distortion types. ? ? ? ? ? ? ? ?.

judicial

A Decision-Theoretic Approach to Forecasting,ior model assumptions made about this process. This chapter describes a general (universal) approach to statistical forecasting based on mathematical decision theory, including a brief discussion of discriminant analysis. The following fundamental notions are introduced: optimal and suboptimal forec

摘要記錄

Time Series Models of Statistical Forecasting,ary time series models, the ARIMA(., ., .) model, nonlinear models, multivariate time series models (including VARMA(., .) and simultaneous equations models), as well as models of discrete time series with a specific focus on high-order Markov chains.

preservative

Performance and Robustness Characteristics in Statistical Forecasting,n the most general form and then specialized for point or interval forecasting and different levels of prior uncertainty. The chapter introduces performance characteristics of forecasting statistics based on loss functions and risk functionals. In order to define mathematically rigorous robustness c

Armada

Soliloquy