Reservoir Computing for Macroeconomic Modelling
Lead Research Organisation:
University College London
Department Name: Statistical Science
Abstract
With the increased globalisation of financial markets, financial crises and shocks have become more frequent and contagious. Despite a mass of data on financial markets and the banking sector being available, traditional macroeconomic models account for financial influences primarily through monetary aggregates and interest rates. This narrow representation of financial markets is proving increasingly inadequate, reflected in poor forecasting performance of traditional macroeconomic models.
A major weakness in the existing models is the absence of aggregate variables which can adequately represent broad financial market conditions. This absence has stimulated a growing literature on the construction of aggregate financial conditions indices (FCIs) formed from a large set of financial variables by use of dynamic factor models (DFMs). Prior research by the PI and Co-I have highlighted some of the limitations of DFM-based FCIs. Such FCIs do not perform well if the number of financial variables considered becomes large and have difficulties taking country specific characteristics into account. Due to these shortcomings, their predictive performance has been disappointing.
The project suggests applying new techniques in machine learning, namely reservoir computing, in the context of macroeconomic modelling and forecasting. Machine learning techniques have shown remarkable forecasting performance in dealing with large datasets, thereby potentially addressing the shortcomings of conventional factor-based FCIs. The project will draw on the expertise of an interdisciplinary and cross-sector research team of economists, statisticians, mathematicians, data and computer scientists and economic practitioners. The PI (a statistician) and Co-I (an economist) will team up with two leading experts in reservoir computing as research partners and the Ministry of finance, Greece as policy partner.
Reservoir computing is a relatively recent approach to machine learning and has not been applied to macroeconomic modelling and forecasting yet. The technique is particularly promising in the given context as it can deal with a mix of time series data and unstructured data, handle frequency mismatch and high dimensionality of input data well, and provides interpretable results, a key condition from an economic modelling perspective. Multiple empirical studies confirm exceptional pertinence of reservoir computing in various forecasting exercises with similar data types as in the macro-finance domain.
The identification of machine learning techniques that can improve macroeconomic modelling and forecasting by use of high dimensional input data that adequately represent financial market dynamics is potentially groundbreaking for economists and economic practitioners. To ensure maximum policy relevance of research outputs, the research team will involve academic beneficiaries and research users in the design, execution and dissemination of the research. All research outputs, including scoping studies, research briefings and journal publications as well as code written in Python, will be made available open source for researchers and research users to access.
The research will be conducted over 12 months in three phases. The first three months are dedicated to involving a wider research community and setting up a strategic network of economists, data and computer scientists, statisticians, mathematicians and economic practitioners. The subsequent six month of the project are dedicated to developing the application of reservoir computing to macroeconomic modelling and forecasting by the project team. The final three months of the project are dedicated to testing and competitively evaluating the performance of new methods in machine learning (including but not limited to reservoir computing) and other fields when applied to macroeconomic modelling and forecasting through the organisation of a 'hackathon' challenge.
A major weakness in the existing models is the absence of aggregate variables which can adequately represent broad financial market conditions. This absence has stimulated a growing literature on the construction of aggregate financial conditions indices (FCIs) formed from a large set of financial variables by use of dynamic factor models (DFMs). Prior research by the PI and Co-I have highlighted some of the limitations of DFM-based FCIs. Such FCIs do not perform well if the number of financial variables considered becomes large and have difficulties taking country specific characteristics into account. Due to these shortcomings, their predictive performance has been disappointing.
The project suggests applying new techniques in machine learning, namely reservoir computing, in the context of macroeconomic modelling and forecasting. Machine learning techniques have shown remarkable forecasting performance in dealing with large datasets, thereby potentially addressing the shortcomings of conventional factor-based FCIs. The project will draw on the expertise of an interdisciplinary and cross-sector research team of economists, statisticians, mathematicians, data and computer scientists and economic practitioners. The PI (a statistician) and Co-I (an economist) will team up with two leading experts in reservoir computing as research partners and the Ministry of finance, Greece as policy partner.
Reservoir computing is a relatively recent approach to machine learning and has not been applied to macroeconomic modelling and forecasting yet. The technique is particularly promising in the given context as it can deal with a mix of time series data and unstructured data, handle frequency mismatch and high dimensionality of input data well, and provides interpretable results, a key condition from an economic modelling perspective. Multiple empirical studies confirm exceptional pertinence of reservoir computing in various forecasting exercises with similar data types as in the macro-finance domain.
The identification of machine learning techniques that can improve macroeconomic modelling and forecasting by use of high dimensional input data that adequately represent financial market dynamics is potentially groundbreaking for economists and economic practitioners. To ensure maximum policy relevance of research outputs, the research team will involve academic beneficiaries and research users in the design, execution and dissemination of the research. All research outputs, including scoping studies, research briefings and journal publications as well as code written in Python, will be made available open source for researchers and research users to access.
The research will be conducted over 12 months in three phases. The first three months are dedicated to involving a wider research community and setting up a strategic network of economists, data and computer scientists, statisticians, mathematicians and economic practitioners. The subsequent six month of the project are dedicated to developing the application of reservoir computing to macroeconomic modelling and forecasting by the project team. The final three months of the project are dedicated to testing and competitively evaluating the performance of new methods in machine learning (including but not limited to reservoir computing) and other fields when applied to macroeconomic modelling and forecasting through the organisation of a 'hackathon' challenge.
Publications
Ballarin G
(2023)
Reservoir computing for macroeconomic forecasting with mixed-frequency data
in International Journal of Forecasting
Description | We introduce a new framework called the Multi-Frequency Echo State Network (MFESN), which originates from a relatively novel machine learning paradigm called reservoir computing (RC) and apply it to macroeconomic forecasting. Specifically, we forecast US GDP growth, using mixed frequency small and medium sized macroeconomic and financial input datasets. We compare the forecasting performance of this new framework with two state-of-the-art time series forecasting methods: MIxed-DAta Sampling (MIDAS) and Dynamic Factor Models (DFM). We find that our MFESN models achieve comparable or better performance than MIDAS and DFMs at a much lower computational cost. |
Exploitation Route | The KOF Nowcasting lab at the ETH at Zurich which is is a real-time platform for testing current and next-quarter GDP estimates in different countries has adopted our methodology and it is currently used in their platform. We are currently considering collaboration with the Bank of England for possible adoption of our method in their portfolio of standard techniques. |
Sectors | Financial Services and Management Consultancy Government Democracy and Justice |
URL | https://www.sciencedirect.com/science/article/pii/S0169207023001085 |
Description | Our methodology is now used in the KOF Nowcasting Lab, in ETH Zurich. The KOF Nowcasting Lab is a real-time platform for testing current and next-quarter GDP estimates in different countries. The growth rate of real gross domestic product (GDP) is an important indicator of a country's economic activity. However, official estimates are published with a lag. In the academic literature there are various methods of forecasting the current quarter's GDP using previously available and higher-frequency data. The KOF Nowcasting Lab is a real-time platform for testing these methods and simultaneously provides an overview of the current economic situation. The models are updated daily for a number of countries based on large amounts of data, and the results are published. |
First Year Of Impact | 2023 |
Sector | Financial Services, and Management Consultancy,Government, Democracy and Justice |
Impact Types | Policy & public services |
Description | 5th Vienna workshop in microeconomics and finance |
Form Of Engagement Activity | Participation in an activity, workshop or similar |
Part Of Official Scheme? | No |
Geographic Reach | International |
Primary Audience | Professional Practitioners |
Results and Impact | Giovanni Ballarin presented our work in this conference. |
Year(s) Of Engagement Activity | 2022 |
URL | https://www.econbiz.de/events/event/5th-vienna-workshop-on-high-dimensional-times-series-in-macroeco... |
Description | A talk on Fintech conference in Milan, Italy |
Form Of Engagement Activity | A talk or presentation |
Part Of Official Scheme? | No |
Geographic Reach | International |
Primary Audience | Industry/Business |
Results and Impact | The PI, Petros Dellaportas, was invited to present the work of this grant to a wide international audience |
Year(s) Of Engagement Activity | 2022 |
URL | https://www.fintechlab.it/fintech_conference2022/ |
Description | Greek stochastics mu |
Form Of Engagement Activity | Participation in an activity, workshop or similar |
Part Of Official Scheme? | No |
Geographic Reach | International |
Primary Audience | Professional Practitioners |
Results and Impact | Petros Dellaportas presented the work of the grant. |
Year(s) Of Engagement Activity | 2022 |
URL | http://www.stochastics.gr/meetings/mu/index.html |
Description | Hybrid workshop at UCL |
Form Of Engagement Activity | Participation in an activity, workshop or similar |
Part Of Official Scheme? | No |
Geographic Reach | International |
Primary Audience | Professional Practitioners |
Results and Impact | All members of the working group presented their work and final project outputs. |
Year(s) Of Engagement Activity | 2022 |
Description | We organised a workshop (on line) |
Form Of Engagement Activity | Participation in an activity, workshop or similar |
Part Of Official Scheme? | No |
Geographic Reach | International |
Primary Audience | Professional Practitioners |
Results and Impact | We organised a workshop to disseminate our initial findings and receive feedback from expert practitioners. |
Year(s) Of Engagement Activity | 2021 |