Forecasting UK inflation bottom up

Joseph, Andreas; Kalamara, Eleni; Kapetanios, George; Potjagailo, Galina

doi:10.2139/ssrn.3819286

Cited by 12 publications

(11 citation statements)

References 34 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…They illustrate this result for US inflation. Similarly, Joseph et al, (2021) show that using disaggregated information improves inflation forecasts for the United Kingdom and Weber and Zika (2016) confirm this result when forecasting German employment. In contrast, the results for the inclusion of disaggregate information in dynamic factor models are mixed (Alvarez et al, 2016).…”

Section: Introductionsupporting

confidence: 53%

Euro Area Income and Wealth Effects: Aggregation Issues*

et al. 2021

View full text Add to dashboard Cite

This empirical study for private consumption in the euro area and the four largest euro area countries validates the need to separate components of income and supports the common practice to split wealth. A focus on income and wealth components avoids an aggregation bias in estimated income and wealth effects. It finds little evidence in favour of an extensive country aggregation issue given direct euro area estimates of income and wealth propensities and forecasts for consumption are not too different from those indirectly aggregated across the four largest euro area countries.JEL Classification numbers: C53, D12, E21, E27. *All views expressed in this paper are those of the authors and do not necessarily reflect the position of the European Central Bank or the Central Bank of Ireland. We thank two anonymous referees and the editor as well as participants of Eurosystem meetings and of various conferences for their valuable comments.

show abstract

Section: Introductionsupporting

confidence: 53%

Euro Area Income and Wealth Effects: Aggregation Issues*

et al. 2021

View full text Add to dashboard Cite

show abstract

“…The dataset contains 112 macroeconomic and financial indicators divided into nine categories: labour, production, retail and services, consumer and retail price indices, producer price indices, international trade, money, credit and interest rate, stock market and finally sentiment and leading indicators. The selection of variables is inspired by McCracken and Ng (2016), Fortin-Gagnon et al (2018) and Joseph et al (2021). The complete list of series is available in table A7.…”

Section: Uk-md: a Large Uk Monthly MDmentioning

confidence: 99%

Can Machine Learning Catch the Covid-19 Recession?

Coulombe

Marcellino

Stevanović

2021

NIER

View full text Add to dashboard Cite

Based on evidence gathered from a newly built large macroeconomic dataset (MD) for the UK, labelled UK-MD and comparable to similar datasets for the United States and Canada, it seems the most promising avenue for forecasting during the pandemic is to allow for general forms of nonlinearity by using machine learning (ML) methods. But not all nonlinear ML methods are alike. For instance, some do not allow to extrapolate (like regular trees and forests) and some do (when complemented with linear dynamic components). This and other crucial aspects of ML-based forecasting in unprecedented times are studied in an extensive pseudo-out-of-sample exercise.

show abstract

“…The third main contribution is to take advantage of recent advancements in ML in order to explain the inflation predictions of these black-box models. This relates to a previous paper by Joseph et al (2021) that deals with black boxes and examines the drivers of forecasts of UK inflation. Our study is the first to focus on explaining the drivers of inflation forecasts for an emerging economy.…”

mentioning

confidence: 88%

“…This involves identifying both the key informative variables but also the sign and the magnitude of their effects on each individual prediction. Our paper differs from that of Joseph et al (2021) in two ways: by including additional ML models in the analysis and by proposing the integration of feature selection methods with ML models to address the problem of model interpretability in high-dimensional settings.…”

mentioning

confidence: 99%