On break-even correlation: the way to price structured credit derivatives by replication

We consider the pricing of European-style structured credit pay-off under the Gaussian Copula Model (GCM). When no sudden jump-to-default events occur, the perfect replication of these pay-offs under the GCM is obtained if and only if the underlying single-name credit spreads follow a particular family of dynamics and if the pricing parameters are given by so-called ‘break-even’ correlations. We exhibit a class of Merton-style models that are consistent with this result. We calculate break-even correlations explicitly to price nth-to-default baskets under the GCM. Finally, we illustrate the usefulness of this concept as a relative-value tool.     

Market volatility and the dynamic hedging of multi-commodity price risk

Commodity cash and futures prices experienced a severe boom-and-bust cycle between 2006 and 2009. Increases in commodity price volatility have raised concerns about the usefulness of commodity futures and options as risk management tools. Dynamic hedging strategies have the potential to improve risk management when conditional (co)variances depart significantly from their unconditional, long-run counterparts and may be useful to decision-makers despite their greater complexity and higher transaction costs. We propose a Nonparametric Copula-based Generalized Autoregressive Conditional Heteroscedastic (NPC-GARCH) approach to estimate time-varying hedge ratios, and evaluate the benefits of dynamic hedging during four sub-periods between 2000 and 2011 using a stylized Texas cattle feedlot management problem. The NPC-GARCH approach allows for a flexible, nonlinear and asymmetric dependence structure between cash and futures prices for different commodities. We find that NPC-GARCH dynamic hedging performs better than either static, GARCH-Dynamic Conditional Correlation (DCC) or GARCH-Baba, Engle, Kraft and Kroner (BEKK) hedging in terms of lower tail risk (expected shortfall), but that there is no significant difference between hedging approaches in terms of portfolio variance reduction.     

Forecasting US real house price returns over 1831–2013: evidence from copula models

Given the existence of nonnormality and nonlinearity in the data generating process of real house price returns over the period of 1831–2013, this article compares the ability of various univariate copula models, relative to standard benchmarks (naive and autoregressive models) in forecasting real US house price over the annual out-of-sample period of 1874–2013, based on an in-sample of 1831–1873. Overall, our results provide overwhelming evidence in favour of the copula models (Normal, Student’s t, Clayton, Frank, Gumbel, Joe and Ali-Mikhail-Huq) relative to linear benchmarks, and especially for the Student’s t-copula, which outperforms all other models both in terms of in-sample and out-of-sample predictability results. Our results highlight the importance of accounting for nonnormality and nonlinearity in the data generating process of real house price returns for the US economy for nearly two centuries of data.     

股票资产组合VaR研究基于混合D藤Copula模型

本文在对上证市场五种股票资产组合的风险分析中以VaR作为风险度量指标,采用基于Pair Copula高维建模理论的混合D藤Copula模型,建立了反应多个资产组合相关结构的联合分布模型。该模型对传统D藤Copula建模方法作了进一步的改进,通过一定的选择标准,确定了D藤中每个Pair Copula函数的最优函数族,这样使得所建立的模型不仅考虑到了资产维数的影响,而且还能捕捉到组合内部因子间相关结构的差异性,从而改进后的模型能更好地描述资产组合的相关结构,并且能更精确地反映资产组合收益的实际分布。最后,以混合D藤Copula模型为基础,利用Monte Carlo方法计算了上证市场五种股票资产组合的VaR,并通过实证研究进一步证明了该模型的有效性。     

Maximizing Profits for a Multi-Category Catalog Retailer

It is a common trend in the retail industry for catalog retailers to mail multiple catalogs, each promoting different product categories. The existing catalog mailing models do not address the issue of optimizing multi-category catalog mailing. We address this research gap by introducing a model that integrates the when and what components of a customer's purchase decision into the how much component (number of catalogs) of a firm's cross-selling strategy. In addition to comparing the impact of category-specific versus full product catalogs in generating sales in a specific category, the study also finds relative impacts of various category-specific catalogs. We jointly estimate the probability of purchase and purchase amounts in multiple product categories by using multivariate proportional hazard model (MVPHM) and a regression based purchase amount model in a Hierarchical Bayesian framework. The model accounts for unobserved heterogeneity, and uses a control function (CF) approach to account for endogeneity in catalog mailing. The results from the Genetic Algorithm (GA) based optimization suggest that the catalog mailing policy as per the proposed model would be able to generate 38.4 percent more customer lifetime value (CLV) from a sample of 10 percent of the households as compared to the current catalog mailing policy of the retailer by reallocation of the catalogs across customers and mailing periods based on their propensity to buy.     

Impact of copula choice on the modeling of crop yield basis risk

A number of problems in agricultural economics involve modeling joint distributions for which the assumption of multivariate normality may not be warranted. Yet, very little work has been conducted evaluating competing methods for modeling joint dependence. We develop a simulation framework to evaluate the bias and efficiency impacts of copula choice in the context of evaluating county‐to‐farm basis risk. The results suggest significant differences in performance across various copulas and approaches. The findings have important implications for risk analysis, insurance, and policy modeling problems in agriculture regarding the selection of method to model dependence among random variables.     

A data-driven approach to forecasting ground-level ozone concentration

The ability to forecast the concentration of air pollutants in an urban region is crucial for decision-makers wishing to reduce the impact of pollution on public health through active measures (e.g. temporary traffic closures). In this study, we present a machine learning approach applied to forecasts of the day-ahead maximum value of ozone concentration for several geographical locations in southern Switzerland. Due to the low density of measurement stations and to the complex orography of the use-case terrain, we adopted feature selection methods instead of explicitly restricting relevant features to a neighborhood of the prediction sites, as common in spatio-temporal forecasting methods. We then used Shapley values to assess the explainability of the learned models in terms of feature importance and feature interactions in relation to ozone predictions. Our analysis suggests that the trained models effectively learned explanatory cross-dependencies among atmospheric variables. Finally, we show how weighting observations helps to increase the accuracy of the forecasts for specific ranges of ozone’s daily peak values.     

Correlations among cryptocurrencies: Evidence from multivariate factor stochastic volatility model

This paper is the first study to apply the multivariate factor stochastic volatility model (MFSVM) for analyzing the correlations among six cryptocurrencies. We use MFSVM with the Bayesian estimation procedure for the period from August 8, 2015, to January 1, 2020. According to the findings, there is a significant positive correlation between price volatility values of Bitcoin and Litecoin. Besides, the volatility values of Ethereum have a positive correlation with both Ripple and Stellar. There is also a positive correlation between the volatility values of Ripple and Dash. These findings are robust to consider different correlation networks. The evidence implies that Bitcoin is mainly related to Litecoin, but Ethereum is associated with other cryptocurrencies.     

多元知识探寻与后发企业技术创新

当后发企业由追随向引领转型时，将致力于对全球化技术、基础科学等多种知识来源的积极探寻，以推进自身技术创新。基于专利及引文数据，分析海外技术知识与科学知识对后发企业技术创新的影响效应。实证研究发现：海外技术知识转移水平与后发企业创新呈U型关系；知识来源的地理多样性则正向作用于后发企业创新。另外，企业科学关联度对后发企业技术创新未起到显著促进作用。在后发转型背景下，明晰企业对多元知识的技术学习机制和效果，对企业创新追赶及其国际化具有重要指导意义。