Home > Research > Publications & Outputs > A joint Bayesian forecasting model of judgment ...

Electronic data

View graph of relations

A joint Bayesian forecasting model of judgment and observed data: Working paper 2012: 4

Research output: Working paper

Published

Standard

A joint Bayesian forecasting model of judgment and observed data : Working paper 2012: 4. / Davydenko, Andrey; Fildes, Robert.

Lancaster : Department of Management Science, Lancaster University, 2012.

Research output: Working paper

Harvard

Davydenko, A & Fildes, R 2012 'A joint Bayesian forecasting model of judgment and observed data: Working paper 2012: 4' Department of Management Science, Lancaster University, Lancaster.

APA

Davydenko, A., & Fildes, R. (2012). A joint Bayesian forecasting model of judgment and observed data: Working paper 2012: 4. Department of Management Science, Lancaster University.

Vancouver

Davydenko A, Fildes R. A joint Bayesian forecasting model of judgment and observed data: Working paper 2012: 4. Lancaster: Department of Management Science, Lancaster University. 2012 Oct.

Author

Davydenko, Andrey ; Fildes, Robert. / A joint Bayesian forecasting model of judgment and observed data : Working paper 2012: 4. Lancaster : Department of Management Science, Lancaster University, 2012.

Bibtex

@techreport{c8fd0cadb6ec48fb81a9f1ae219471cc,
title = "A joint Bayesian forecasting model of judgment and observed data: Working paper 2012: 4",
abstract = "This paper presents a new approach that aims to incorporate prior judgmental forecasts into a statistical forecasting model. The result is a set of forecasts that are consistent with both the judgment and latest observations. The approach is based on constructing a model with a combined dataset where the expert forecasts and the historical data are described by means of corresponding regression equations. Model estimation is done using numeric Bayesian analysis. Semiparametric methods are used to ensure finding adequate forecasts without any prior knowledge of the specific type of the trend function. The expert forecasts can be provided as estimates of future time series values or as estimates of total or average values over any particular time intervals. Empirical analysis has shown that the approach is operable in practical settings. Compared to standard methods of combining, the approach is more flexible and in empirical comparisons proves to be more accurate.",
keywords = "Forecasting accuracy, combining statistical methods and judgement",
author = "Andrey Davydenko and Robert Fildes",
year = "2012",
month = oct,
language = "English",
publisher = "Department of Management Science, Lancaster University",
type = "WorkingPaper",
institution = "Department of Management Science, Lancaster University",

}

RIS

TY - UNPB

T1 - A joint Bayesian forecasting model of judgment and observed data

T2 - Working paper 2012: 4

AU - Davydenko, Andrey

AU - Fildes, Robert

PY - 2012/10

Y1 - 2012/10

N2 - This paper presents a new approach that aims to incorporate prior judgmental forecasts into a statistical forecasting model. The result is a set of forecasts that are consistent with both the judgment and latest observations. The approach is based on constructing a model with a combined dataset where the expert forecasts and the historical data are described by means of corresponding regression equations. Model estimation is done using numeric Bayesian analysis. Semiparametric methods are used to ensure finding adequate forecasts without any prior knowledge of the specific type of the trend function. The expert forecasts can be provided as estimates of future time series values or as estimates of total or average values over any particular time intervals. Empirical analysis has shown that the approach is operable in practical settings. Compared to standard methods of combining, the approach is more flexible and in empirical comparisons proves to be more accurate.

AB - This paper presents a new approach that aims to incorporate prior judgmental forecasts into a statistical forecasting model. The result is a set of forecasts that are consistent with both the judgment and latest observations. The approach is based on constructing a model with a combined dataset where the expert forecasts and the historical data are described by means of corresponding regression equations. Model estimation is done using numeric Bayesian analysis. Semiparametric methods are used to ensure finding adequate forecasts without any prior knowledge of the specific type of the trend function. The expert forecasts can be provided as estimates of future time series values or as estimates of total or average values over any particular time intervals. Empirical analysis has shown that the approach is operable in practical settings. Compared to standard methods of combining, the approach is more flexible and in empirical comparisons proves to be more accurate.

KW - Forecasting accuracy

KW - combining statistical methods and judgement

M3 - Working paper

BT - A joint Bayesian forecasting model of judgment and observed data

PB - Department of Management Science, Lancaster University

CY - Lancaster

ER -