A Bayesian Indicator of Manufacturing Output from Qualitative Business Panel Survey Data
Martin Weale, James Mitchell
Abstract
Open-access reader
Martin Weale, James Mitchell
Abstract
Open-access reader
Qualitative business survey data are used widely to provide indicators of economic activity ahead of the publication of official data. Traditional indicators exploit only aggregate survey information, namely the proportions of respondents who report "up" and "down". This paper considers disaggregate or firm-level survey responses. It derives an alternative Bayesian indicator of economic activity relating firms' categorical responses to official data using ordered discrete-choice models. An application to firm-level survey data from the Confederation of British Industry shows that the proposed indicator of manufacturing output growth can provide more accurate early estimates of manufacturing output growth than traditional indicators.
OpenAlex reports 5 citations for this work. Citation counts describe recorded attention and do not establish research quality.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
Qualitative business survey data are used widely to provide indicators of economic activity ahead of the publication of official data. Traditional indicators exploit only aggregate survey information, namely the proportions of respondents who report "up" and "down". This paper considers disaggregate or firm-level survey responses. It derives an alternative Bayesian indicator of economic activity relating firms' categorical responses to official data using ordered discrete-choice models. An application to firm-level survey data from the Confederation of British Industry shows that the proposed indicator of manufacturing output growth can provide more accurate early estimates of manufacturing output growth than traditional indicators.
Key concepts: Categorical variable, Survey data collection, Exploit, Bayesian probability, Econometrics, Aggregate (composite), Aggregate data, Economic indicator