Estimation of Cyclical Time Series: an Application to Central England Temperatures

Seminar summary

In this paper we present a methodology that corrects for the irregular space between realizations of unevenly spaced time series and provides accurate estimates of AR model parameters. This methodology is simple in its application and its asymptotic properties are investigated thoroughly using Monte Carlo simulations. Furthermore, using results for estimation of AR parameters under missing data, we develop a successful time series tool for estimation of time series with cyclically varying parameters. More specifically, we assume periodicity of the mean and autoregressive parameters of such models. To illustrate the robustness and flexibility of the method, we conduct an application, modelling daily temperatures data. This estimation technique allows us to uncover cyclical (daily as annual) patterns in the data without imposing restrictive assumptions. Using the Central England Temperature (CET) time series (1772 - today) and the daily England-Wales precipitation series (1931 - today) we find with a high level of accuracy that temperature intra-year average and persistence have increased in the sample 1850-2020 compared to 1772 - 1850, especially for the winter months, while the variance of the random shocks of AR(1) model fitted to the data seems to be decreasing over time.

How to join this seminar

This seminar will take place on Wednesday 15 November 2023 at 3pm.

We warmly welcome you to join us in person in EBS.2.66, Essex Business School Building, Colchester Campus.

If you are unable to make it to us the seminar will be available to watch live on Zoom. Please contact Dr Ilias Chronopoulos for the passcode.

Speaker bio

Fulvia Marotta

Fulvia Marotta is a postdoctoral research associate at the University of Oxford, with a PhD in Economics and a background in Statistics. Her research interests lie at the intersection of time series climate econometrics, applied macroeconomics, and environmental economics.

Fulvia's work revolves around the use of time series and macro-econometric methods to understand how climate change impacts economies and how countries can become more resilient to compound shocks. Her research focuses on the role of fiscal policy in building long-term resilience and sustainable development, and how green fiscal policy can be integrated with immediate crisis responses to fortify a country's resilience. Fulvia has projects investigating the effect of climate change on energy and commodity prices, the optimal combination of environmental policy and green innovation for a net-zero economy, and the implications of policy-induced shifts in green innovation on firms and labour productivity.

Fulvia has authored several working papers in the fields of econometrics and climate change and has presented her research at many conferences and seminars, including the NBER summer institute and those hosted by the European Central Bank and other national central banks. Visit Fulvia's personal website for more information.