Forecasting with exponential smoothing : the state space by Rob J. Hyndman ... [et al.]

By Rob J. Hyndman ... [et al.]

Exponential smoothing equipment were round because the Fifties, and are the most well-liked forecasting equipment utilized in company and undefined. This booklet brings jointly numerous effects at the nation house framework for exponential smoothing. it truly is of curiosity to humans eager to practice the tools of their personal region of interest.

Basic techniques --
Getting began --
Linear concepts kingdom house types --
Nonlinear and heteroscedastic options kingdom house types --
Estimation of concepts nation area versions --
Prediction distributions and durations --
Selection of types --
Normalizing seasonal parts --
Models with regressor variables --
Some homes of linear versions --
Reduced types and relationships with ARIMA versions --
Linear techniques country house types with random seed states --
Conventional country house versions --
Time sequence with a number of seasonal styles --
Nonlinear types for optimistic information --
Models for count number information --
Vector exponential smoothing --
Inventory keep an eye on functions --
Conditional heteroscedasticity and purposes in finance --
Economic functions : the Beveridge-Nelson decomposition --
References --
Author index --
Data index --
Subject index

Show description

Read or Download Forecasting with exponential smoothing : the state space approach PDF

Best game theory books

Quests: Design, Theory, and History in Games and Narratives

This precise tackle quests, incorporating literary and electronic concept, presents a very good source for online game builders. serious about either the idea and perform of the 4 major points of quests (spaces, items, actors, and challenges), every one theoretical part is by means of a pragmatic part that comprises workouts utilizing the Neverwinter Nights Aurora Toolset.

Queueing Theory: A Linear Algebraic Approach

Queueing conception bargains with platforms the place there's rivalry for assets, however the calls for are just recognized probabilistically. This publication will be regarded as both a monograph or a textbook at the topic, and hence is aimed toward audiences. it may be invaluable in case you already understand queueing conception, yet wish to comprehend extra in regards to the linear algebraic technique.

Advances in Mathematical Economics Volume 20

The sequence is designed to assemble these mathematicians who're heavily attracted to getting new difficult stimuli from financial theories with these economists who're looking potent mathematical instruments for his or her study. loads of monetary difficulties should be formulated as restricted optimizations and equilibration in their suggestions.

Extra resources for Forecasting with exponential smoothing : the state space approach

Sample text

1b) where r (·) and w(·) are scalar functions, f (·) and g (·) are vector functions, and ε t is a white noise process with variance σ2 . , when the series contains only non-negative values). Nevertheless, the Gaussian assumption is often a reasonable approximation when the level of the process is sufficiently far from the origin (or, more generally, the region of impossible values) and it will then be convenient to use the Gaussian assumption as a basis for inference. The functions in this model may all be time-indexed, but we shall concentrate on constant functions (the invariant form), albeit with time-varying arguments.

8) j =0 where dt = w F t−1 x0 , k0 = 1 and k j = w F j−1 g for j = 1, 2, . . Thus, the observation is a linear function of the seed state x0 and past and present errors. 8); this is an example of a finite Wold decomposition (Brockwell and Davis 1991, p. 180). 38 3 Linear Innovations State Space Models The model is described as stationary2 if ∞ ∑ |k j | < ∞ and j =0 lim dt = d. 8) converge rapidly to zero, and the impact of the seed state vector diminishes over time. We may then consider the limiting form of the model, corresponding to the infinite start-up assumption.

Specifically, a forecastable model has the properties ∞ ∑ |c j | < ∞ j =1 and lim at = a. 3 Model Properties 37 Our definition of forecastability allows the initial state x0 to have an ongoing effect on forecasts, but it prevents observations in the distant past having any effect. In most cases, a = 0, but not always; an example with a = 0 is given in Sect. 2. 7) to hold is that the eigenvalues of D lie inside the unit circle. In this case, D j converges to a null matrix as j increases. This is known as the “stability condition” and such models are called stable.

Download PDF sample

Rated 4.53 of 5 – based on 21 votes