International Journal of Financial Engineering and Risk Management (13 papers in press)
An Equity-Credit Hybrid Model for Asset Correlations
by Fabio Dias
Abstract: Single factor Gaussian copula models are widely used to manage credit risk of loan portfolios, even driving how many large financial institutions are capitalised under Basel II / III. Under this formulation, the default correlation between two separate firms is directly explained by their asset correlation to a systematic factor, which can be estimated using either equity correlations or observed default rates, with the portfolio losses usually being simulated under a Gaussian copula model. Though it is widely accepted that the use of observed default rates or even equity returns to calibrate a single factor Gaussian copula model is likely to understate the tail risk, this paper proposes a Bayesian approach for a single factor Gaussian copula where the asset correlations are modelled using an inverse Wishart prior with the scale parameter calibrated to observed default rates and the degrees of freedom chosen using the in-sample continuous ranked probability score whilst the equity correlations are used to obtain the posterior distribution. The proposed hybrid model is shown to produce probabilistic forecasts of defaults with better out-of-sample performance than the standard single factor Gaussian copula even though it maintained low complexity and ease of implementation.
Keywords: asset correlations; credit risk management; structural model of credit risk; factor copula models.
Special Issue on: Behavioural Finance and Decision-Making in Financial Markets
Trading the Stock Market using Google Search Volumes
by Joseph St. Pierre, Mateusz Klimkiewicz, Adonay Resom, Nikolaos Kalampalikis
Abstract: In this paper, we present a methodology for utilizing Google Search Indices obtained from the Google Trends website as a means for measuring potential investor interest in stocks listed on the Dow Jones Index (Dow 30). We accomplish this task by utilizing a Long Short-Term Memory network that correlates changes in the search volume for a given asset with changes in the actual trade volume for said asset. Additionally, by using these predictions, we formulate a concise trading strategy in the hopes of being able to outperform the market and analyze the results of this new strategy by backtesting across weekly closing price data for the last six months of 2016. With these tests, it was discovered that about 43% of the time, the machine learning based trading strategy outperformed the baseline sample indicating that there is indeed a correlation between price movements for certain assets on the Dow 30 and the number of Google searches for said assets. Furthermore, while the scope of our study was limited to the Dow 30 in order to mitigate selection bias, we nonetheless hypothesize that numerous other assets that similarly possess a predictable correlation to Google Search Volumes are likely to exist thereby making the trading algorithm described in this paper applicable beyond the narrow scope of this study.
Keywords: Machine Learning; Stock Prediction; Market Prediction; Artificial Intelligence; DOW 30;.
Towards an improved Credit Scoring System: the Greek case.
by Christos Kountzakis
Abstract: During the development of credit risk assessment models, it is very important to find variables that allow the evaluation of a companys credit risk accurately, as the classification results depend on the appropriate characteristics for a selected data set. In this paper, new credit risk models tested on real data, which evaluate credit risk of Greek companies are introduced. These models use a combination of financial and credit behavior data. The credit risk models, which are introduced in this paper do have some important additional advantages: a) they contain a relatively small number of variables, b) their stability is tested on samples after the time-period of the timeperiod of data-collection and c) the characterization of 'good' and 'bad' credit behavior is strictly defined.
Keywords: credit risk; logistic regression; financial ratios.
Racial Discrimination in TARP Investments
by Lucas Puente, Linus Wilson
Abstract: Minority and black owned banks were significantly less likely to receive funds from the Troubled Asset Relief Program (TARP) Community Development Capital Initiative (CDCI). A non-minority bank with the median characteristics was approximately ten times more likely to obtain TARP funds than an African American owned bank after controlling for other factors. We also find prior TARP recipients and banks with fewer troubled assets were more likely to obtain money from this program.
Keywords: Barney Frank; bailout; CDCI; CDFI; Community Development Capital Initiative; Community Development Financial Institution; discrimination; ethics; EESA; Emergency Economic Stabilization Act; Maxine Waters; minority ownership; OneUnited bank; politics; preferred stock; race; racial discrimination; subordinated debt; U.S. House Financial Services Committee; TARP.
Determinants of Mortgage Arrears: The case of Buy to Let
by Alexios Makropoulos
Abstract: This paper considers an error correction modelling (VECM) approach to identify determinants of mortgage portfolio arrears in the Buy to Let (BTL) segment of the UK mortgage market. Within this context, the alternative theories of the ability to pay view and the equity view are tested for the particular segment of the UK mortgage market. The empirical results suggest that UK Buy to Let arrears are related to the dual-trigger approach where elements of the ability to pay view are combined with elements from the equity view to determine the aggregate arrears behaviour. From a behavioural perspective the results imply that borrowers decisions may be influenced from personal biases and optimism around the long-run affordability and equity returns of their (BTL) investment decisions. Likewise, the 2008 crisis provided some indication that lenders may be affected from similar biases when making lending decisions. These findings may therefore be of practical use for mortgage portfolio managers and decision makers when considering policies for the BTL segment of the UK mortgage market.
Keywords: Credit risk; Buy to Let; BTL; mortgage arrears; mortgage defaults.
The dividend policy of tourist firms pre and during the economic crisis in Greece
by Mihail Diakomihalis, Nikolaos Kapsiohas
Abstract: The hotel industry constitutes one of the most significant branches of the Greek economy and is the fundamental pillar of Greek tourism. Hotel enterprises have not suffered notably from the financial crisis, thus demonstrating that tourism has been the heavy industry of Greece. In this paper, an extended survey was conducted regarding companies with share capital, their way of taxation, and also the way they dispose of their profits. Particular emphasis was given to corporations (S.A., or limited liability companies). Subsequently, S.A. companies were ex-amined on the basis of their turnover and the region of Greece where they are located. Finally, twenty hotel S.A. companies in highly developed tourist regions were selected.
The analysis was based on their published financial data covering the period 2004-2014. It was confirmed that in hotel enterprises as in all enterprises in Greece taxation was in-creased, especially in 2013 and 2014. Nevertheless, this increase does not seem to have had a great impact on distributed dividends to stockholders.
Furthermore, it is of particular interest that the turnover rate is low in all enterprises. It is an interesting point since it indicates that these enterprises, even though the vast majority of them raised both their capital and fixed assets, appear to have overinvested funds with respect to the amount of their sales, despite the upward turnover in 2014 and throughout the years of economic crisis in general.
The healthy status of Greek hotel enterprises is proven by the growth of their current assets, since 80% significantly increased their current assets, 15% maintained their current assets sta-ble and only 0.05% experienced a decrease.
Keywords: Tourist enterprises; dividends; taxation; economic crisis.
Quantitative Easing and Government Bonds: Evidence from the EU
by Antonios Sarantidis, Fotios Mitropoulos, Konstantinos Kollias
Abstract: This paper examines the relationship of quantitative easing policy programs, government bond yields and banking stock price returns for the EU periphery. The aim is to estimate and to analyze the effect of the ECBs QE policy on government bond yields and banking stock prices. The dataset consist of monthly panel data observations spanning from January 2008 until September 2017. The empirical part utilizes the Difference-in-Differences methodology. The findings suggest that the implementation of QE has positive effects, by decreasing the bond yields for the periphery countries that participated in the program; there were no effects noted for banking returns. Additional macroeconomic variables are included, where a reduction in government debt and unemployment and a raise in industrial production, tax revenues and FDI lead to lower government bond yields and to higher banking returns. The research provides to investors and government regulators empirical insights into the effects that QE policies have.
Keywords: Quantitative easing; Government bonds; Banking returns; Difference-in-Difference.
Trading CSR/CSE Leveraged Inefficiency
by Vasiliki Basdekidou
Abstract: The paper is about the inefficiency (market anomaly) appearing in leveraged trading instruments, like 3x equity ETFs, with CSR/CSE firms as benchmark index. The subject is important because, nowadays, leveraged ETFs have grown in popularity, in particular within the intraday trading community, because they could generate returns very quickly (right side of the trade). The principal target is to introduce a framework for personalised market Entry/Exit tactics when trading leveraged ETFs. For this purpose, the concept CSR/CSE leverage inefficiency is defined initially as a innovative term benchmarked a 3-d array, and then the dimensions, functionalities, and trading returns of this inefficiency (as the core part of a proposed trading methodology), is discussed. Papers findings state that, in sideways and choppy markets overnight-position institutions profit from the proposed concept at the expense of long-term investors and swing traders as well. Similarly in trending markets, day-trading speculators profit at the expense of hedge-funds. Main papers contribution is the empirically-tested conclusion that, after the incorporation of the CSR/CSE leverage inefficiency in trading tactics: (i) in choppy markets, institutional money accumulates profit entirely with overnight-positions in ethical non-leveraged ETFs; while (ii) in a trending market, the profit occurs in day-trading on non-ethical leveraged instruments traded mainly by speculators. Finally, according to a comparative return analysis performing by this article, the best results in all cases are received by CSR/CSE ETFs in sideways and choppy markets after employing an overnight-position return trading strategy.
Keywords: CSR; CSE; ETF; leverage; market anomaly; trading; market volatility; trading functionalities; trending markets; sideways markets.
Special Issue on: Applications of Optimisation in Finance
Why Your Smart Beta Portfolio Might Not Work
by Yongjae Lee, Woo Chang Kim
Abstract: Smart beta, which accounts for rule-based factor-tilting strategies that fall between active and passive investment, has emerged as an alternative to active investment after its major decline since the global financial crisis. In spite of the smart betas remarkable commercial prosperity, many experts in both industry and academia share some concerns. Some of them believe that the marketing hype might confuse investors while others are concerned about the exposure to unintended risks that smart beta products might bring. In this study, we provide a comprehensive review of diverse perspectives from both practitioners and researchers on smart beta, and we perform empirical and theoretical investigations on the efficiency of smart beta (or factor-tilting) strategies as investment building blocks. We find that factor-based investment building blocks may cause inefficiency under the mean-variance framework.
Keywords: Smart beta; factor investing; investment building block; passive investment.
Asset-liability management and Goal-based Investing for Retail Business
by Giorgio Consigli, Massimo Di Tria
Abstract: The industry of online personal financial services is expected over the next years to absorb an increasing share of households' and individuals' savings and investment decisions with a parallel expansion of tailored decision tools and underlying methodological developments. In this article we extend our previous work on long-term retirement planning and present a dynamic stochastic optimization model formulated to tackle an optimal wealth management problem based explicitly on the introduction of consumption and investment goals with a terminal inflation-adjusted retirement target over a long-term horizon. By embedding a goal-based investing philosophy in a dynamic framework we extend our previous results and provide a reference modeling approach for increasingly popular households asset-liability management services. In a discrete setting we show that a dynamic stochastic programming formulation will lead to a highly realistic representation and solution of an otherwise hardly manageable optimization problem and it is consistent with computer-aided decision support tools' operational requirements.
Keywords: dynamic stochastic programming; households’ finance; asset-liability management; goal-based investing; life-cycle.
Special Issue on: Applications of Optimisation in Finance
Machine Learning, Economic Regimes and Portfolio Optimization
by John M. Mulvey, Han Hao, Nongchao Li
Abstract: In portfolio models, the depiction of future outcomes depends upon a representative accounting of economic conditions. There is much evidence that crash periods display much different patterns than normal markets, suggesting that forecasting models ought to be based on multiple regimes. We apply two techniques from machine learning in our empirical study to improve robustness: 1) trend filtering - to distinguish regimes possessing relatively homogeneous patterns, and 2) a shrinkage/cross validation approach within a factor analysis of performance. A scenario based portfolio model is proposed and designed to address multiple regimes. The worst-case events are well described within the framework, as compared with historical performance.
Keywords: Portfolio models; asset allocation; economic regimes; machine learning; classification,factor investing.
Factor-Based Optimization and the Creation/Redemption Mechanism of Fixed Income Exchange-Traded Funds
by Bennett Golub, Maurizio Ferconi, Ananth Madhavan, Alex Ulitsky
Abstract: Fixed income exchange-traded funds (ETFs) trade on organized exchanges, often with narrow spreads, liquidity, and pre- and post-trade transparency. The remarkable success of fixed income ETFs relies critically on the efficient functioning of the ETF creation-redemption mechanism which helps drive the funds market price to stay closely in line with the underlying values of the bond portfolios they represent. Creation of fixed income ETFs faces challenges though because of the less liquid nature of the markets for many bonds. In this article, we explain how custom fixed income baskets can be used with exchange-traded funds in a systematic, auditable, and repeatable manner. We use factor-based optimization to create ETF baskets for one or more ETFs and with one or many counter-parties. We conclude that optimization can improve the efficiency of ETF basket creation, which in turn induces higher liquidity and tighter spreads, benefitting investors.
Keywords: Fixed Income; ETFs; Factors.
Multi-period portfolio optimization with alpha decay
by Kartik Sivaramakrishnan, Dieter Vandenbussche
Abstract: The traditional Markowitz MVO approach is based on a single-period model. Single period models do not utilize any data or decisions beyond the rebalancing time horizon with the result that their policies are "myopic" in nature. For long-term investors, multi-period optimization offers the opportunity to make"wait-and-see" policy decisions by including approximate forecasts and long-term policy decisions beyond the rebalancing time horizon. We consider portfolio optimization with a composite alpha signal that is composed of a short-term and a long-term alpha signal. The short-term alpha better predicts returns at the end of the rebalancing period but it decays quickly, i.e., it has less memory of its previous values. On the other hand, the long-term alpha has less predictive power than the short-term alpha but it decays slowly. We develop a simple two stage multi-period model that incorporates this alpha model to construct the optimal portfolio at the end of the rebalancing period. We compare this model with the traditional single-period MVO model on a simulated example from Israelov & Katz and also a large strategy with realistic constraints and show that the multi-period model tends to generate portfolios that are likely to have a better realized performance.
Keywords: Markowitz mean-variance optimization; Multi-period portfolio optimization; Alpha decay.