Mark Cummins

The research question we consider is whether incremental complexity in option pricing models is justified by incremental model performance. We apply the model confidence set as a formal model comparison approach in appraising stochastic volatility jump‐diffusion option pricing models, spanning affine and nonaffine specifications. Jumps in price with stochastic (constant) arrival intensity produce superior (inferior) outcomes. A parsimonious negative exponential price jump distribution outperforms the popular normal distribution. Jumps in volatility (synchronized or not) worsen model performance. A parsimonious nonlinear hyperbolic drift extension of the Heston model performs particularly well. Nonlinear CEV models generally do not produce appreciable model performance.

Globalisation, technological advances, liberalisation of financial markets, and changing consumer behaviour are transforming banking profoundly. Under the EU Payment Services Directive 2 (PSD2), incumbent banks must open up their data, processes, and business functionalities to customers and third parties including rivals. It is critical to understand consumer behaviour post-PSD2, and the potential impact of PSD2 on the functioning of the retail banking and financial services market. In this preliminary study of 244 consumers from six European countries, we explore the role of social influence, facilitating conditions, perceived risk, and effort and performance expectancy in order to unravel the determinants of consumers’ acceptance of account information services (AIS) as provisioned under PSD2, which provide consolidated bank account information for consumers with multiple bank accounts across multiple banking institutions. Our findings suggest that the competing influences of (a) positive perceptions such as social influence, facilitating conditions, and performance expectancy, and (b) negative perceptions related to risk, sway consumers’ intentions to adopt AIS.

An objective of model validation within organisations is to provide guidance on model selection decisions that balance the operational effectiveness and structural complexity of competing models. We consider a practice-relevant model validation scenario where a financial quantitative analysis team seeks to decide between incumbent and alternative models on the basis of parameter risk. We devise a model risk management methodology that gives a meaningful distributional assessment of parameter risk in a setting where market calibration and historical estimation procedures must be jointly applied. Such a scenario is typically driven by data constraints that preclude market calibration only. We demonstrate our proposed methodology in a natural gas storage modelling context, where model usage is necessary to support profit and loss reporting, and to inform trading and hedging strategy. We leverage our distributional parameter risk approach to devise an accessible technique to support model selection decisions.

In the context of data that is missing completely at random, we propose a new data imputation method that exploits Cholesky decomposition. The data imputation method falls within the multiple imputation framework and is designed to ensure consistency with the correlation structure of the available data. The advantage is an accessible and computationally efficient approach to managing missing data that avoids the model risk associated with applying complex model-based data imputation methods. The non-recursive nature of our data imputation method further avoids the convergence issues associated with recursive approaches.

Alleviation of information asymmetries is a major challenge in online credit markets. We provide new insights into the lending decisions of non-bank financial institutions in this setting by analysing all loans issued by the largest peer-to-peer lending platform in the UK over a three year period. We benchmark the lending performance of institutional investors against that of retail investors, collectively referred to as ‘the crowd’. We test the performance of institutions in screening loan opportunities, examining ex-ante loan rates against ex-post loan default, and we also analyse loan payoff through realised returns. We find that institutions outperform the crowd, as a collective, across probability of loan default, loan rates charged, and ex-post gains on repaid loans, although institutions lose substantially more upon loan default. In accounting for sample selection bias, we find no evidence that the funding decisions of institutions show a pattern of minimising loan default risk. In identifying potential adverse selection and moral hazard in smaller loans and unsecured loans, we find that institutions have a preference for larger loans. Formally we show that institutions are less exposed to default risk on small loan lending, while being more exposed on large loan lending. Segmenting the crowd by size, a more nuanced interpretation of our initial conclusion emerges. Institutions only outperform small and medium sized crowds, with comparable performance against the lending decisions of large sized crowds. Finally, we note that institutions are able to exploit the auction process to their advantage relative to small and medium sized crowds, although this disappears when the platform moves to its current fixed rate system. Our study provides important insights into the lending decisions of non-bank financial institutions in online credit markets, with implications for their future lending activity.

Psychological barriers are shown to be a feature in frontier market equity pricing. Using MSCI Frontier 100 index constituents; prices, upon breaching an upward or downward psychologically important price point, generally follow predictable patterns in the days after such a breach. It is demonstrated that certain characteristics of frontier markets drive the presence of this relationship. There are regional effects, and also cultural influences as measured by country individualism scores. Liquidity also appears to mediate the presence of psychological barriers. This constitutes a first study of behavioral influences in these fast-growing markets that are being increasingly recognised as an asset class with strong diversification benefits for investors previously restricted to developed and emerging equity markets.

(2016). Model Risk in Financial Markets: From Financial Engineering to Risk Management. Quantitative Finance: Vol. 16, No. 9, pp. 1333-1337.