LAMBDA Research Centre

This theme investigates key aspects of firms’ strategic decisions and their aggregate economic implications using large micro datasets, answering questions, such as how firms:

• Price their products and compete in the global and local markets
• Manage their supply chains and interact with their business partners
• Row by upgrading their technologies or expanding to new markets.

A key goal is to understand how answers to these questions change under different market conditions and economic shocks.

Recent research in this area focuses on the impact of COVID-19, studying firms’ price and production decisions during the outbreak, investigating how firms’ responses to the crisis differ depending on their characteristics and quantifying the aggregate implications of these responses.

Within this theme, we investigate and evaluate the impacts of monetary, fiscal, trade and industrial policies on economic outcomes.

Recent policies and events analysed include the impact of training programs and grants on firm and labour market outcomes, the effects of innovation policies, as well as recent trade policies and the effects of Brexit on UK exports.

Such research leads to substantial impact by evaluating specific policies and formulating recommendations for policy makers.

Thanks to the colossal size of financial data and growing computational power, applying machine learning techniques in finance is becoming more popular and rigorous.  

Our projects under this theme include, but are not limited to:

• Portfolio management using estimation
• Error minimisation via machine learning
• Algorithmic trading exploring systematic miss-pricing via machine learning
• Account information fraud detection using textual analysis and machine learning
• High dimensional risk management using machine learning.

One important domain in which big data techniques can be applied to is business and management, as insights derived from its analysis can assist different size firms and policymakers in designing well informed strategies and accurately assessing business outcomes.

This often incudes big data sources at various operational levels, such as sales, cost, R&D expenses, etc.

In addition to data obtained from firms’ daily operation, for some statistical simulation exercises, key parameters collected from the real world can also be implemented to enlighten critical business decisions.

Within this theme, we actively engage in the forefront of theoretical and methodological research for modern econometric and big data analyses.

We have internationally leading and excellent research expertise and publish in areas broadly including:

• Bayesian and machine learning methods
• Semi- and non-parametric techniques for high dimensional or high frequency data
• Time series analysis techniques for modelling structural stability, regime switching, dynamic dependence, as well as continuous time stochastic processes and count data.