Página 19 dos resultados de 4919 itens digitais encontrados em 0.012 segundos

## Measuring Forecasters' Perceptions of Inflation Persistence

Fonte: Quens University
Publicador: Quens University

Tipo: Tese de Doutorado

EN; EN

Relevância na Pesquisa

26.66%

#forecast revisions#conservatism#inflation expectations#rational bias#Survey of Professional Forecasters#inflation persistence

This dissertation presents a new measure of U.S. inflation persistence from the point of view of a professional forecaster. In chapter 2 I explore two different measures that give insight into the views of professional forecasters and link their views with U.S. inflation data. One of these measures, given by the persistence implied by forecast revisions, appears to have similarities with actual inflation persistence over the 1981–2008 sample period.
Chapter 3 explores forecast revisions in a more general setting allowing forecasters to have their own views on inflation persistence as well as a unique information set. This chapter builds a measure of perceived inflation persistence via the implied autocorrelation function that follows from the estimates obtained using a forecaster-specific state-space model. When compared to the autocorrelation function for actual inflation, forecasters tend to react less to shocks that hit inflation than the actual inflation data would suggest. This could be due to increased credibility of the Federal Reserve, but it could also be a result of a bias in the underlying inflation forecasts.
Chapter 4 focuses on this issue and finds that the reluctance of forecasters to make revisions to their previously announced forecasts causes their estimates of perceived inflation persistence to be understated as their announced inflation forecasts differ from their true inflation expectations. This chapter also presents a method to undo this bias by retrieving their true inflation expectations series.; Thesis (Ph.D...

Link permanente para citações:

## Proyección de demanda: ¡este problema no es normal!

Fonte: Universidad Icesi; Facultad de Ciencias Administrativas y Económicas; Administración de Empresas; Departamento de Gestión Organizacional
Publicador: Universidad Icesi; Facultad de Ciencias Administrativas y Económicas; Administración de Empresas; Departamento de Gestión Organizacional

Tipo: article; Artículo
Formato: pdf; 237-239 páginas; Digital

SPA

Relevância na Pesquisa

26.66%

#Demanda#Producción intelectual registrada - Universidad Icesi#Modelos de regresión#Predicción#Linear regression model#Forecast#Semi-logarithmic model#Normal error term

El objetivo de este caso es discutir sobre la aproximación adecuada para realizar predicciones cuando se emplea un modelo de regresión lineal con una forma semilogarítmica. Para ello, se contextualiza el problema de una firma que requiere realizar predicciones sobre las cantidades demandadas de su producto estrella. De esta manera, el lector se enfrenta al desafío de cuestionar si una aproximación aparentemente intuitiva genera pronósticos adecuados.; This case discusses a problem that may arise when using a semi-logarithmic linear model to forecast the demand of a given product. This problem is presented using the discussion that emerges in a company that is looking for an approach to forecast the demand of its leading product. The reader faces the challenge of deciding which approach is adequate to the problem.

Link permanente para citações:

## Optimizing inventory levels using financial, lifecycle and forecast variance data

Fonte: Massachusetts Institute of Technology
Publicador: Massachusetts Institute of Technology

Tipo: Tese de Doutorado
Formato: 48 leaves

ENG

Relevância na Pesquisa

26.66%

#Sloan School of Management.#Electrical Engineering and Computer Science.#Leaders for Manufacturing Program.

Significant inventory write-offs have recently plagued ATI Technologies, a world leader in graphics and media processors. ATI's product-centric culture has long deterred attention from supply chain efficiency. Given that manufacturing lead time exceeds customer order lead time for its semiconductors, ATI relies heavily on their demand forecasting team to instigate supply chain activities. The PC business unit forecasting team translates market information into product-line forecast and also sets finished goods inventory levels intended to offset demand uncertainty. Today's inventory decisions are made in response to customer escalations, often ignoring financial implications. To add necessary rigor when setting these inventory levels, this thesis presents a model using wafer and unit cost, profit margin, product lifecycle stage and historical forecast error to categorize products into inventory risk levels. The resultant risk levels become a critical input to monthly demand-supply meetings with marketing, operations and senior executives - the outcome of which are wafer orders and assembly and test plans at the world's largest contract foundries and subcontractors. Finally, the 2006 acquisition of ATI by Advanced Micro Devices (AMD) offers unforeseen flexibility...

Link permanente para citações:

## Estudo empírico sobre as notas colocadas em circulação em Portugal

Fonte: Instituto Superior de Economia e Gestão
Publicador: Instituto Superior de Economia e Gestão

Tipo: Dissertação de Mestrado

Publicado em //2015
POR

Relevância na Pesquisa

26.66%

Mestrado em Econometria Aplicada e Previsão; Com a introdução do euro a análise do uso e circulação efetiva de notas por cada um dos países da zona euro passou a ser uma tarefa bastante desafiante devido à sua contínua migração entre países. É também cada vez mais importante a estimação e previsão da evolução da circulação de notas estando o DET encarregue de assegurar a emissão e a colocação em circulação da moeda legal necessária à economia nacional. O objetivo deste estudo é auxiliar o DET na sua tarefa e fazer um exercício comparativo da capacidade preditiva de um conjunto de modelos econométricos clássicos para prever a circulação de notas em Portugal, através de uma abordagem econométrica. Numa primeira fase é feita uma análise sobre a relação entre as notas em circulação e um conjunto de variáveis económicas através da abordagem VEC para o período entre 2002 e 2014. Os resultados demonstram evidência estatística de existência de uma relação de cointegração, havendo portanto uma relação de equilíbrio de longo prazo entre as variáveis. Numa segunda fase são testadas várias metodologias de previsão para avaliar qual o modelo com melhor desempenho para prever as notas em circulação. Tanto os três critérios para avaliar os erros de previsão como o teste Diebold Mariano apontaram para o método de Holt como aquele com melhor capacidade preditiva sobre as notas em circulação. Relativamente à previsão para o período de 2015 até 2017...

Link permanente para citações:

## Oceanic stochastic parametrizations in a seasonal forecast system

Fonte: Universidade Cornell
Publicador: Universidade Cornell

Tipo: Artigo de Revista Científica

Publicado em 30/06/2015

Relevância na Pesquisa

26.66%

We study the impact of three stochastic parametrizations in the ocean
component of a coupled model, on forecast reliability over seasonal timescales.
The relative impacts of these schemes upon the ocean mean state and ensemble
spread are analyzed. The oceanic variability induced by the atmospheric forcing
of the coupled system is, in most regions, the major source of ensemble spread.
The largest impact on spread and bias came from the Stochastically Perturbed
Parametrization Tendency (SPPT) scheme - which has proven particularly
effective in the atmosphere. The key regions affected are eddy-active regions,
namely the western boundary currents and the Southern Ocean. However, unlike
its impact in the atmosphere, SPPT in the ocean did not result in a significant
decrease in forecast error. Whilst there are good grounds for implementing
stochastic schemes in ocean models, our results suggest that they will have to
be more sophisticated. Some suggestions for next-generation stochastic schemes
are made.; Comment: 24 pages, 3 figures

Link permanente para citações:

## Using Regression Kernels to Forecast A Failure to Appear in Court

Fonte: Universidade Cornell
Publicador: Universidade Cornell

Tipo: Artigo de Revista Científica

Publicado em 05/09/2014

Relevância na Pesquisa

26.66%

Forecasts of prospective criminal behavior have long been an important
feature of many criminal justice decisions. There is now substantial evidence
that machine learning procedures will classify and forecast at least as well,
and typically better, than logistic regression, which has to date dominated
conventional practice. However, machine learning procedures are adaptive. They
"learn" inductively from training data. As a result, they typically perform
best with very large datasets. There is a need, therefore, for forecasting
procedures with the promise of machine learning that will perform well with
small to moderately-sized datasets. Kernel methods provide precisely that
promise. In this paper, we offer an overview of kernel methods in regression
settings and compare such a method, regularized with principle components, to
stepwise logistic regression. We apply both to a timely and important criminal
justice concern: a failure to appear (FTA) at court proceedings following an
arraignment. A forecast of an FTA can be an important factor is a judge's
decision to release a defendant while awaiting trial and can influence the
conditions imposed on that release. Forecasting accuracy matters, and our
kernel approach forecasts far more accurately than stepwise logistic
regression. The methods developed here are implemented in the R package kernReg
currently available on CRAN.; Comment: 43 pages...

Link permanente para citações:

## Forecast of the Chemical Aging and Relevant Color Changes in Painting

Fonte: Universidade Cornell
Publicador: Universidade Cornell

Tipo: Artigo de Revista Científica

Publicado em 05/05/2005

Relevância na Pesquisa

26.66%

The article describes the potential application of thermodynamic simulation
to forecast chemical aging and relevant color changes in painting. Qualitative
and numerical results were obtained by applying the method to various mixtures
of pigments without and with atmospheric components. The results were compared
to the legendary recommendations on incompatible pigment mixtures with about an
80 percent match regarding potential color changes in the aged mixtures.
Results for the cadmium yellow-lead white and cadmium lemon-emerald green
mixtures are illustrated by pictures, gradually showing color changes caused by
the aging. The method of thermodynamic simulation can be a powerful tool to
investigate old masterpieces, in developing new materials, and to forecast some
aspects of the aging of real masterpieces.; Comment: 7 pages, 3 tables, 2 figures

Link permanente para citações:

## MSSM Forecast for the LHC

Fonte: Universidade Cornell
Publicador: Universidade Cornell

Tipo: Artigo de Revista Científica

Relevância na Pesquisa

26.66%

We perform a forecast of the MSSM with universal soft terms (CMSSM) for the
LHC, based on an improved Bayesian analysis. We do not incorporate ad hoc
measures of the fine-tuning to penalize unnatural possibilities: such
penalization arises from the Bayesian analysis itself when the experimental
value of $M_Z$ is considered. This allows to scan the whole parameter space,
allowing arbitrarily large soft terms. Still the low-energy region is
statistically favoured (even before including dark matter or g-2 constraints).
Contrary to other studies, the results are almost unaffected by changing the
upper limits taken for the soft terms. The results are also remarkable stable
when using flat or logarithmic priors, a fact that arises from the larger
statistical weight of the low-energy region in both cases. Then we incorporate
all the important experimental constrains to the analysis, obtaining a map of
the probability density of the MSSM parameter space, i.e. the forecast of the
MSSM. Since not all the experimental information is equally robust, we perform
separate analyses depending on the group of observables used. When only the
most robust ones are used, the favoured region of the parameter space contains
a significant portion outside the LHC reach. This effect gets reinforced if the
Higgs mass is not close to its present experimental limit and persits when dark
matter constraints are included. Only when the g-2 constraint (based on
$e^+e^-$ data) is considered...

Link permanente para citações:

## The climate version of the Eta regional forecast model. 1. Evaluation of consistency between the Eta model and HadAM3P global model

Fonte: Universidade Cornell
Publicador: Universidade Cornell

Tipo: Artigo de Revista Científica

Relevância na Pesquisa

26.66%

The regional climate model prepared from Eta WS forecast model has been
integrated over South America with the horizontal resolution of 40 km for the
period of 1960-1990. The model was forced at its lateral boundaries by the
outputs of HadAM3P. The data of HadAM3P represent simulation of modern climate
with the resolution about 150 km. In order to prepare climate regional model
from the Eta forecast model multiple modifications and corrections were made in
the original model. The run of climate Eta model was made on the supercomputer
SX-6. The detailed analysis of the results of dynamical downscaling experiment
includes an investigation of a consistency between the regional and AGCM models
as well as of ability of the regional model to resolve important features of
climate fields on the finer scale than that resolved by AGCM. In this work the
results of the investigation of a consistency between the output fields of the
Eta model and HadAM3P are presented. The geopotential, temperature and wind
fields of both models were analysed. For the evaluation of the likeness of
these two models outputs,there were used Fourier analysis of time series,
consistency index, constituted from linear regression coefficients, time mean
and space mean models' arithmetic difference and root mean square difference...

Link permanente para citações:

## Parameter uncertainty in forecast recalibration

Fonte: Universidade Cornell
Publicador: Universidade Cornell

Tipo: Artigo de Revista Científica

Relevância na Pesquisa

26.66%

Ensemble forecasts of weather and climate are subject to systematic biases in
the ensemble mean and variance, leading to inaccurate estimates of the forecast
mean and variance. To address these biases, ensemble forecasts are
post-processed using statistical recalibration frameworks. These frameworks
often specify parametric probability distributions for the verifying
observations. A common choice is the Normal distribution with mean and variance
specified by linear functions of the ensemble mean and variance. The parameters
of the recalibration framework are estimated from historical archives of
forecasts and verifying observations. Often there are relatively few forecasts
and observations available for parameter estimation, and so the fitted
parameters are also subject to uncertainty. This artefact is usually ignored.
This study reviews analytic results that account for parameter uncertainty in
the widely used Model Output Statistics recalibration framework. The predictive
bootstrap is used to approximate the parameter uncertainty by resampling in
more general frameworks such as Non-homogeneous Gaussian Regression. Forecasts
on daily, seasonal and annual time scales are used to demonstrate that
accounting for parameter uncertainty in the recalibrated predictive
distributions leads to probability forecasts that are more skilful and reliable
than those in which parameter uncertainty is ignored. The improvements are
attributed to more reliable tail probabilities of the recalibrated forecast
distributions.; Comment: 22 pages...

Link permanente para citações:

## Forecast and event control: On what is and what cannot be possible

Fonte: Universidade Cornell
Publicador: Universidade Cornell

Tipo: Artigo de Revista Científica

Relevância na Pesquisa

26.66%

Consequences of the basic and most evident consistency requirement-that
measured events cannot happen and not happen at the same time-are shortly
reviewed. Particular emphasis is given to event forecast and event control. As
a consequence, particular, very general bounds on the forecast and control of
events within the known laws of physics are derived. These bounds are of a
global, statistical nature and need not affect singular events or groups of
events.; Comment: Paper presented on the Workshop on "Determinism" at Ringberg Castle,
Germany, July 6th, 2001

Link permanente para citações:

## A new method for choosing parameters in delay reconstruction-based forecast strategies

Fonte: Universidade Cornell
Publicador: Universidade Cornell

Tipo: Artigo de Revista Científica

Relevância na Pesquisa

26.66%

#Computer Science - Information Theory#Mathematics - Dynamical Systems#Physics - Data Analysis, Statistics and Probability

Delay-coordinate reconstruction is a proven modeling strategy for building
effective forecasts of nonlinear time series. The first step in this process is
the estimation of good values for two parameters, the time delay and the
embedding dimension. Many heuristics and strategies have been proposed in the
literature for estimating these values. Few, if any, of these methods were
developed with forecasting in mind, however, and their results are not optimal
for that purpose. Even so, these heuristics---intended for other
applications---are routinely used when building delay coordinate
reconstruction-based forecast models. In this paper, we propose a new strategy
for choosing optimal parameter values for forecast methods that are based on
delay-coordinate reconstructions. The basic calculation involves maximizing the
shared information between each delay vector and the future state of the
system. We illustrate the effectiveness of this method on several synthetic and
experimental systems, showing that this metric can be calculated quickly and
reliably from a relatively short time series, and that it provides a direct
indication of how well a near-neighbor based forecasting method will work on a
given delay reconstruction of that time series. This allows a practitioner to
choose reconstruction parameters that avoid any pathologies...

Link permanente para citações:

## Planetary Kp index forecast using autoregressive models

Fonte: Universidade Cornell
Publicador: Universidade Cornell

Tipo: Artigo de Revista Científica

Publicado em 10/04/2014
ES

Relevância na Pesquisa

26.66%

The geomagnetic Kp index is derived from the K index measurements obtained
from thirteen stations located around the Earth geomagnetic latitudes between
$48^\circ$ and $63^\circ$. This index is processed every three hours, is
quasi-logarithmic and estimates the geomagnetic activity. The Kp values fall
within a range of 0 to 9 and are organized as a set of 28 discrete values. The
data set is important because it is used as one of the many input parameters of
magnetospheric and ionospheric models. The objective of this work is to use
historical data from the Kp index to develop a methodology to make a prediction
in a time interval of at least three hours. Five different models to forecast
geomagnetic indices Kp and ap are tested. Time series of values of Kp index
from 1932 to 15/12/2012 at 21:00 UT are used as input to the models. The
purpose of the model is to predict the three measured values after the last
measured value of the Kp index (it means the next 9 hours values). The AR model
provides the lowest computational cost with satisfactory results. The ARIMA
model is efficient for predicting Kp index during geomagnetic disturbance
conditions. This paper provides a quick and efficient way to make a prediction
of Kp index, without using satellite data. Although it is reported that the
forecast results are better when satellite data are used. In the literature we
find that the linear correlation between predicted values and actual values is
$77\%$...

Link permanente para citações:

## The extended Baryon Oscillation Spectroscopic Survey (eBOSS): a cosmological forecast

Fonte: Universidade Cornell
Publicador: Universidade Cornell

Tipo: Artigo de Revista Científica

Publicado em 28/10/2015

Relevância na Pesquisa

26.66%

We present a science forecast for the eBOSS survey, part of the SDSS-IV
project, which is a spectroscopic survey using multiple tracers of large-scale
structure, including luminous red galaxies (LRGs), emission line galaxies
(ELGs) and quasars (both as a direct probe of structure and through the
Ly-$\alpha$ forest). Focusing on discrete tracers, we forecast the expected
accuracy of the baryonic acoustic oscillation (BAO), the redshift-space
distortion (RSD) measurements, the $f_{\rm NL}$ parameter quantifying the
primordial non-Gaussianity, the dark energy and modified gravity parameters. We
also use the line-of-sight clustering in the Ly-$\alpha$ forest to constrain
the total neutrino mass. We find that eBOSS LRGs ($0.60.6$), ELGs ($0.6

Link permanente para citações:

## Forecast of surface layer meteorological parameters at Cerro Paranal with a mesoscale atmospherical model

Fonte: Universidade Cornell
Publicador: Universidade Cornell

Tipo: Artigo de Revista Científica

Publicado em 07/04/2015

Relevância na Pesquisa

26.66%

#Astrophysics - Instrumentation and Methods for Astrophysics#Physics - Atmospheric and Oceanic Physics

This article aims at proving the feasibility of the forecast of all the most
relevant classical atmospherical parameters for astronomical applications (wind
speed and direction, temperature) above the ESO ground-base site of Cerro
Paranal with a mesoscale atmospherical model called Meso-Nh. In a precedent
paper we have preliminarily treated the model performances obtained in
reconstructing some key atmospherical parameters in the surface layer 0-30~m
studying the bias and the RMSE on a statistical sample of 20 nights. Results
were very encouraging and it appeared therefore mandatory to confirm such a
good result on a much richer statistical sample. In this paper, the study was
extended to a total sample of 129 nights between 2007 and 2011 distributed in
different parts of the solar year. This large sample made our analysis more
robust and definitive in terms of the model performances and permitted us to
confirm the excellent performances of the model. Besides, we present an
independent analysis of the model performances using the method of the
contingency tables. Such a method permitted us to provide complementary key
informations with respect to the bias and the RMSE particularly useful for an
operational implementation of a forecast system.; Comment: 20 pages...

Link permanente para citações:

## Large Scale Structure Forecast Constraints on Particle Production During Inflation

Fonte: Universidade Cornell
Publicador: Universidade Cornell

Tipo: Artigo de Revista Científica

Relevância na Pesquisa

26.66%

#Astrophysics - Cosmology and Nongalactic Astrophysics#General Relativity and Quantum Cosmology#High Energy Physics - Phenomenology#High Energy Physics - Theory

Bursts of particle production during inflation provide a well-motivated
mechanism for creating bump like features in the primordial power spectrum.
Current data constrains these features to be less than about 5% the size of the
featureless primordial power spectrum at wavenumbers of about 0.1 h Mpc^{-1}.
We forecast that the Planck cosmic microwave background experiment will be able
to strengthen this constraint to the 0.5% level. We also predict that adding
data from a square kilometer array (SKA) galaxy redshift survey would improve
the constraint to about the 0.1% level. For features at larger wave-numbers,
Planck will be limited by Silk damping and foregrounds. While, SKA will be
limited by non-linear effects. We forecast for a Cosmic Inflation Probe (CIP)
galaxy redshift survey, similar constraints can be achieved up to about a
wavenumber of 1 h Mpc^{-1}.; Comment: 10 pages. Matches PRD accepted version

Link permanente para citações:

## Earthquake Forecast via Neutrino Tomography

Fonte: Universidade Cornell
Publicador: Universidade Cornell

Tipo: Artigo de Revista Científica

Relevância na Pesquisa

26.66%

We discuss the possibility of forecasting earthquakes by means of
(anti)neutrino tomography. Antineutrinos emitted from reactors are used as a
probe. As the antineutrinos traverse through a region prone to earthquakes,
observable variations in the matter effect on the antineutrino oscillation
would provide a tomography of the vicinity of the region. In this preliminary
work, we adopt a simplified model for the geometrical profile and matter
density in a fault zone. We calculate the survival probability of electron
antineutrinos for cases without and with an anomalous accumulation of electrons
which can be considered as a clear signal of the coming earthquake, at the
geological region with a fault zone, and find that the variation may reach as
much as 3% for $\bar \nu_e$ emitted from a reactor. The case for a $\nu_e$ beam
from a neutrino factory is also investigated, and it is noted that, because of
the typically high energy associated with such neutrinos, the oscillation
length is too large and the resultant variation is not practically observable.
Our conclusion is that with the present reactor facilities and detection
techniques, it is still a difficult task to make an earthquake forecast using
such a scheme, though it seems to be possible from a theoretical point of view
while ignoring some uncertainties. However...

Link permanente para citações:

## MOSE: optical turbulence and atmospherical parameters operational forecast at ESO ground-based sites. II: atmospherical parameters in the surface layer [0-30] m

Fonte: Universidade Cornell
Publicador: Universidade Cornell

Tipo: Artigo de Revista Científica

Publicado em 26/09/2013

Relevância na Pesquisa

26.66%

#Astrophysics - Instrumentation and Methods for Astrophysics#Physics - Atmospheric and Oceanic Physics#Physics - Data Analysis, Statistics and Probability

This article is the second of a series of articles aiming at proving the
feasibility of the forecast of all the most relevant classical atmospherical
parameters for astronomical applications (wind speed and direction,
temperature, relative humidity) and the optical turbulence (Cn2 and the derived
astro-climatic parameters like seeing, isoplanatic angle, wavefront coherence
time...). This study is done in the framework of the MOSE project, and focused
above the two ESO ground-bases sites of Cerro Paranal and Cerro Armazones. In
this paper we present the results related to the Meso-Nh model ability in
reconstructing the surface layer atmospherical parameters (wind speed
intensity, wind direction and absolute temperature, [0-30] m a.g.l.). The model
reconstruction of all the atmospherical parameters in the surface layer is very
satisfactory. For the temperature, at all levels, the RMSE (Root Mean Square
Error) is inferior to 1{\deg}C. For the wind speed, it is ~2 m/s, and for the
wind direction, it is in the range [38-46{\deg}], at all levels, that
corresponds to a RMSE_relative in a range [21-26{\deg}]. If a filter is applied
for the wind direction (the winds inferior to 3 m/s are discarded from the
computations), the wind direction RMSE is in the range [30-41{\deg}]...

Link permanente para citações:

## MOSE: optical turbulence and atmospherical parameters operational forecast at ESO ground-based sites. I: Overview and atmospherical parameters vertical stratification on [0-20] km

Fonte: Universidade Cornell
Publicador: Universidade Cornell

Tipo: Artigo de Revista Científica

Publicado em 26/09/2013

Relevância na Pesquisa

26.66%

#Astrophysics - Instrumentation and Methods for Astrophysics#Physics - Atmospheric and Oceanic Physics#Physics - Data Analysis, Statistics and Probability

We present the overview of the MOSE project (MOdeling ESO Sites) aiming at
proving the feasibility of the forecast of the classical atmospherical
parameters (wind speed intensity and direction, temperature, relative humidity)
and the optical turbulence OT (CN2 profiles and the most relevant integrated
astro-climatic parameters derived from the CN2: the seeing, the isoplanatic
angle, the wavefront coherence time) above the two ESO ground-based sites of
Cerro Paranal and Cerro Armazones. The final outcome of the study is to
investigate the opportunity to implement an automatic system for the forecast
of these parameters at these sites. In this paper we present results related to
the Meso-Nh model ability in reconstructing the vertical stratification of the
atmospherical parameters along the 20 km above the ground. The very
satisfactory performances shown by the model in reconstructing most of these
parameters (and in particular the wind speed) put this tool of investigation as
the most suitable to be used in astronomical observatories to support AO
facilities and to calculate the temporal evolution of the wind speed and the
wavefront coherence time at whatever temporal sampling. The further great
advantage of this solution is that such estimates can be available in advance
(order of some hours) with respect to the time of interest; Comment: 21 pages...

Link permanente para citações:

## Forecasting Distributions with Experts Advice

Fonte: Universidade de Cambridge
Publicador: Universidade de Cambridge

Formato: 432791 bytes; application/pdf; application/pdf

EN_GB

Relevância na Pesquisa

26.66%

#Classification-JEL: C53, C14#Expert, Forecast Combination, Multiplicative Update, Non-asymptotic Bound, On-line Learning, Shifting.

This paper considers forecasts of the distribution of data whose distribution function is possibly time varying. The forecast is achieved via time varying combinations of experts? forecasts. We derive theoretical worse case bounds for general algorithms based on multiplicative updates of the combination weights. The bounds are useful to study the properties of forecast combinations when data are nonstationary and there is no unique best model. An application with an empirical study is used to highlight the results in practice.

Link permanente para citações: