SSRN Viewer

Can an agent learn efficiently in a noisy and self adapting environment with sequential, non-stationary and non-homogeneous observations? Through trading bots, we illustrate how Deep Reinforcement Learning (DRL) can tackle this challenge. Our contributions are threefold: (i) the use of contextual information also referred to as augmented state in DRL, (ii) the impact of a one period lag between observations and actions that is more realistic for an asset management environment, (iii) the implementation of a new repetitive train test method called walk forward analysis, similar in spirit to cross validation for time series. Although our experiment is on trading bots, it can easily be translated to other bot environments that operate in sequential environment with regime changes and noisy data. Our experiment for an augmented asset manager interested in finding the best portfolio for hedging strategies shows that AAMDRL achieves superior returns and lower risk.

Interest rate swaps are an actively traded product in the financial marketplace and are popular for hedging mortgage and corporate loan exposures against rises in interest rates. Asset swaps on the other hand provide a form of asset financing, where investors borrow funds to purchase an asset, typically a bond. Asset swaps are also a good bond rich-cheap analysis tool. Both types of swaps can of course be used for speculative purposes. In this paper we provide an overview of both interest rate swaps and asset swaps, we explain the products and examine how they are priced & quoted in the market. Analytical and numerical risk is also considered. We conclude with a review of swap pricing formulas and examine how to price swaps quickly in one's head. We do this using simple approximations that hold extremely well in the current low interest rate environment.

Due to their low trading costs, ETFs are potentially a catalyst for short-horizon liquidity traders. The liquidity shocks can propagate to the underlying securities through the arbitrage channel, and ETFs may increase the non-fundamental volatility of the securities in their baskets. We exploit exogenous changes in index membership and find that stocks with higher ETF ownership display significantly higher volatility. ETF ownership increases the negative autocorrelation in stock prices. The increase in volatility appears to introduce undiversifiable risk in prices, because stocks with high ETF ownership earn a significant risk premium of up to 56 basis points monthly.

Non-standard errors capture uncertainty due to differences in research design choices. We establish considerable variation in the design choices made by researchers when constructing asset pricing factors. By purposely data mining over two thousand different versions of each factor, we find that Sharpe ratios vary substantially within a factor due to different construction choices, which results in sizable non-standard errors. We provide simple suggestions that reduce the average non-standard error by 70%. Our study has important implications for model selection exercises.

This paper proposes a practical way of estimating the cost of risky debt for use in the cost of capital. The cost of debt is different from both the promised yield and the risk-free rate, which are sometimes used for this purpose, because of the expected probability of default. The Merton (1974) model of risky debt is employed to decompose the promised yield spread into expected default and return premium components. The advantage of the proposed approach is that all inputs are easily observable. The parameters of the Merton model implied by these inputs are used to compute the expected return on debt. It is argued that, although Merton's framework is simple and stylised, it can be used to estimate the expected return as a fraction of the observed promised market yield in a way consistent with equilibrium. The cost of debt is computed for parameter values that are typical for high grade and low grade debt. It is found that, while using the promised yield as the cost of debt may be adequate for high grade debt, it is likely to cause significant errors for high-yield bonds. In such cases the approach proposed in this paper can be used to adjust the WACC for the probability of default on the firm's debt.

In a previous article we highlighted how traditional stochastic volatility and Jump/Lévy models impose structural constraints on how the short forward skew, the spot/vol correlation, and the term structure of the vol-of-vol are related. Here we propose a model that enables them to be controlled separately and also prices options on realized variance consistently. We present pricing examples for a reverse cliquet, a Napoleon, an accumulator and an option on variance.

We document a striking pattern in U.S. and international stock returns: double sorting on the previous month’s return and share turnover reveals significant short-term reversal among low-turnover stocks, whereas high-turnover stocks exhibit short-term momentum. Short-term momentum is as profitable and as persistent as conventional price momentum. It survives transaction costs and is strongest among the largest, most liquid, and most extensively covered stocks. Our results are difficult to reconcile with models imposing strict rationality but are suggestive of an explanation based on some traders underappreciating the information conveyed by prices.

Short-term alpha signals are generally dismissed in traditional asset pricing models, primarily due to market friction concerns. However, this paper demonstrates that investors can obtain a significant net alpha by applying a combination of signals to a liquid global universe and with advanced buy/sell trading rules that mitigate transaction costs. The composite model consists of short-term reversal, short-term momentum, short-term analyst revisions, short-term risk, and monthly seasonality signals. The resulting alpha is present in out-of-sample and post-publication periods, across regions, translates into long-only applications, is robust to incorporating implementation lags of several days, and is uncorrelated to traditional Fama-French factors.

These presentation slides have been written for the Advanced Course in Asset Management (theory and applications) given at the University of Paris-Saclay. They contain 5 lectures (Part 1. Portfolio Optimization Part 2. Risk Budgeting Part 3. Smart Beta, Factor Investing and Alternative Risk Premia Part 4. Green and Sustainable Finance, ESG Investing and Climate Risk Part 5. Machine Learning in Asset Management) and 15 tutorial exercises.The Table of contents is the following:Part 1. Portfolio Optimization 1. Theory of portfolio optimization 1.a. The Markowitz framework 1.b. Capital asset pricing model (CAPM) 1.c. Portfolio optimization in the presence of a benchmark 1.d. Black-Litterman model2. Practice of portfolio optimization 2.a. Covariance matrix 2.b. Expected returns 2.c. Regularization of optimized portfolios 2.d. Adding constraints3. Tutorial exercises 3.a. Variations on the efficient frontier 3.b. Beta coefficient 3.c. Black-Litterman modelPart 2. Risk Budgeting 1. The ERC portfolio 1.a. Definition 1.b. Special cases 1.c. Properties 1.d. Numerical solution2. Extensions to risk budgeting portfolios 2.a. Definition of RB portfolios 2.b. Properties of RB portfolios 2.c. Diversification measures 2.d. Using risk factors instead of assets3. Risk budgeting, risk premia and the risk parity strategy 3.a. Diversified funds 3.b. Risk premium 3.c. Risk parity strategies 3.d. Performance budgeting portfolios4. Tutorial exercises 4.a. Variation on the ERC portfolio 4.b. Weight concentration of a portfolio 4.c. The optimization problem of the ERC portfolio 4.d. Risk parity fundsPart 3. Smart Beta, Factor Investing and Alternative Risk Premia1. Risk-based indexation 1.a. Capitalization-weighted indexation 1.b. Risk-based portfolios 1.c. Comparison of the four risk-based portfolios 1.d. The case of bonds2. Factor investing 2.a. Factor investing in equities 2.b. How many risk factors? 2.c. Construction of risk factors 2.d. Risk factors in other asset classes3. Alternative risk premia 3.a. Definition 3.b. Carry, value, momentum and liquidity 3.c. Portfolio allocation with ARP4. Tutorial exercises 4.a. Equally-weighted portfolio 4.b. Most diversified portfolio 4.c. Computation of risk-based portfolios 4.d. Building a carry trade exposurePart 4. Green and Sustainable Finance, ESG Investing and Climate Risk 1. ESG investing 1.a. Introduction to sustainable finance 1.b. ESG scoring 1.c. Performance in the stock market 1.d. Performance in the corporate bond market2. Climate risk 2.a. Introduction to climate risk 2.b. Climate risk modeling 2.c. Regulation of climate risk 2.d. Portfolio management with climate risk3. Sustainable financing products 3.a. SRI Investment funds 3.b. Green bonds 3.c. Social bonds 3.d. Other sustainability-linked strategies4. Impact investing 4.a. Definition 4.b. Sustainable development goals (SDG) 4.c. Voting policy, shareholder activism and engagement 4.d. The challenge of reporting5. Tutorial exercises 5.a. Probability distribution of an ESG score 5.b. Enhanced ESG score & tracking error controlPart 5. Machine Learning in Asset Management1. Portfolio optimization 1.a. Standard optimization algorithms 1.b. Machine learning optimization algorithms 1.c. Application to portfolio allocation2. Pattern learning and self-automated strategies3. Market generators4. Tutorial exercises 4.a. Portfolio optimization with CCD and ADMM algorithms 4.b. Regularized portfolio optimization

The advent of the Great Recession in 2008 was the culmination of a perfect storm of lax regulation, a growing housing bubble, rising popularity of derivatives instruments, and questionable banking practices. In addition to these causes, management incentives, as well as certain US accounting standards, contributed to the financial crisis. We outline the significant effects of these incentive structures, and the role of fair value accounting standards during the crisis, and discuss implications and relevance of these rules to practitioners, standard-setters, and academics.

We propose a new investor sentiment index that is aligned with the purpose of predicting the aggregate stock market. By eliminating a common noise component in sentiment proxies, the new index has much greater predictive power than existing sentiment indices both in- and out-of-sample, and the predictability becomes both statistically and economically significant. In addition, it outperforms well recognized macroeconomic variables and can also predict cross-sectional stock returns sorted by industry, size, value, and momentum. The driving force of the predictive power appears stemming from investors’ biased belief about future cash flows.

Spanish Abstract: En esta monografía se describen dos tipos de productos financieros derivados y sus mercados: futuros y opciones. En concreto: la cámara de compensación, las características de ambos activos, la paridad entre el precio del futuro y el de contado, los factores que determinan el valor de una opción, el método binomial de valoración de opciones, el modelo de Black & Scholes, una introducción a la cobertura de riesgo con opciones, otros tipos de opciones (divisas, índices y futuros) así como las opciones exóticas y las opciones día cero. English Abstract: In this monograph two kind of financial derivative assets are analyzed: Futures and options. We study: their main characteristics, the clearing house, the parity between future and spot prices, what factors determine the option value, the binomial method, the Black & Scholes model, an introduction to the risk hedging with options, other kind of options (currency, index and futures), exotic options and zero-day options.

This is the second in a series of articles that examines the practical applications of economic thought. Its focus is on the most fundamental aspects of finance theory, namely asset pricing. We discuss the major pricing models developed during the past 5 decades and critically examine their practical applications. Sadly, the results are not very encouraging. As with other academic economic disciplines, the gap between what is taught about the markets and what actually takes place is quite large, a gap in no way mitigated by the behavioralist arm of the subject. The seminal works of Sharpe and Lintner have provided us with a sound foundation upon which to build realistic pricing models, but unfortunately the unwavering acceptance of these models has resulted in research that merely cements their acceptance, discouraging an examination of how those pricing models could be adapted to suit the practical world.

Deep reinforcement learning (DRL) has reached super human levels in complex tasks like game solving (Go, StarCraft II, Atari Games), and autonomous driving. However, it remains an open question whether DRL can reach human level in applications to financial problems and in particular in detecting pattern crisis and consequently dis-investing. In this paper, we present an innovative DRL framework consisting in two subnetworks fed respectively with portfolio strategies past performances and standard deviations as well as additional contextual features. The second sub network plays an important role as it captures dependencies with common financial indicators features like risk aversion, economic surprise index and correlations between assets that allows taking into account context based information. We compare different network architectures either using layers of convolutions to reduce network’s complexity or LSTM block to capture time dependency and whether previous allocations is important in the modeling. We also use adversarial training to make the final model more robust. Results on test set show this approach substantially over-performs traditional portfolio optimization methods like Markovitz and is able to detect and anticipate crisis like the current COVID one.

I quantitatively measure the nature of the media's interactions with the stock market using daily content from a popular Wall Street Journal column. I find that high media pessimism predicts downward pressure on market prices followed by a reversion to fundamentals, and unusually high or low pessimism predicts high market trading volume. These results and others are consistent with theoretical models of noise and liquidity traders. However, the evidence is inconsistent with theories of media content as a proxy for new information about fundamental asset values, as a proxy for market volatility, or as a sideshow with no relationship to asset markets.

Abstract, by Craig William French This paper reprints an edited version of Jack Treynor's famous 1962 manuscript. The author's facsimile of the original mimeograph was obtained thanks to the kind generosity of Professor Elroy Dimson of the London Business School. Edits in the present version, which differ from the original Rough Draft, include minor typographical corrections and minor notation differences for some variables in the formulae. Specifically, where Mr. Treynor used a bar or curve over a variable, I use an underscore, and I have omitted the upper and lower limits (and their index) above and below all Sigma summation signs as the nature of the summations is clear. Pagination is as in the original. I have otherwise attempted, in this replica, to preserve the vintage character of the document, including font style, spacing, margins and a reproduction reminiscent of the days prior to the invention of photocopy machines and modern portfolio theory. Unfortunately, I could not replicate the wonderful mimeo aroma. Mr. Treynor points out that many minor changes will be evident to anyone who compares the Craig French version to the version in Robert Korajczyk's book, Asset Pricing and Portfolio Performance, Risk Books, London, 1999 [Chapter 2, pp. 15-22]. Mr. Treynor considers all French's changes improvements. A more complete description of the development of the Treynor CAPM may be found in French, Craig W., “The Treynor Capital Asset Pricing Model”. Journal of Investment Management, Vol. 1, No. 2, pp. 6072, 2003. Available at SSRN: http://ssrn.com/abstract=447580

This is a PDF file of 'A Literature Review of Risk Perception Studies in Behavioral Finance: The Emerging Issues' slides from a presentation at the 25th Annual Meeting of the Society for the Advancement of Behavioral Economics (SABE) Conference hosted by New York University on May 15-18, 2007. The topic of risk perception is a noteworthy theme within the behavioral finance literature that examines the decision making process of unsophisticated and expert investors. The risk perception literature has a strong historical academic foundation that entails various attributes associated with the notion of the interdisciplinary and multidisciplinary perspective across the different fields such as behavioral finance, behavioral accounting, and psychology. The previous narrative risk perception literature review by Ricciardi (2004) demonstrated that scholars in financial psychology (behavioral finance), behavioral economics, and behavioral accounting have investigated and tested over 150 unique accounting, financial, and investment proxy risk factors (e.g., beta, current ratio) and more than 100 behavioral risk characteristics (e.g., overconfidence, familiarity bias). This presentation is a preliminary discussion that builds on the research work of Ricciardi (2004, 2006) and Ricciardi (2008). This presentation disclosed, 'What are the emerging issues within the behavioral finance risk perception literature?' In particular, this presentation provided an emerging collection of hypotheses based on the various theories, concepts, and themes from the earlier narrative literature review by Ricciardi (2004) and the forthcoming book chapter in Ricciardi (2008). Note: SSRN is experimenting with enabling the distribution of different types of files: slides, spreadsheets, video, etc. This is an upload of a PDF file of PowerPoint slides. We are interested in our users desires to distribute files that go beyond word processing text files. You can communicate with me on these issues via my email address below. We invite you to submit your own presentation slides.

This paper examines the proportion of wealth invested in stock and bond portfolios as a function of the investors' age, i.e., investment horizon. It has become increasingly popular to advice investors to relocate their funds from a primarily stock portfolio to a primarily bond portfolio as they get older. However, the existing theory does not support this advice: the well-known decision rules such as Mean-Variance (MV) or Stochastic Dominance (SD) rules are unable to explain this common practice. In this paper, we utilize the recently developed Almost Stochastic Dominance (ASD) and Almost Mean Variance (AMV) approaches and employ various datasets to examine the dominance of stock and bond portfolios as a function of the investment horizon. We find that, for short investment horizons, all portfolios are efficient. However, for medium and longer horizons, only the portfolios with higher stock proportions are efficient. The results indicate that ASD and AMV rules unambiguously support the popular practice of advising higher stock to bond ratio for long investment horizons. Hence, we provide an explanation to the practitioners' recommendation within the expected utility paradigm.

We show that short interest is arguably the strongest known predictor of aggregate stock returns. It outperforms a host of popular return predictors both in and out of sample, with annual r-squared statistics of 12.89% and 13.24%, respectively. In addition, short interest can generate utility gains of over 300 basis points per annum for a mean-variance investor. A vector autoregression decomposition shows that the economic source of short interest’s predictive power stems predominantly from a cash flow channel. Overall, our evidence indicates that short sellers are informed traders who are able to anticipate future aggregate cash flows and associated market returns. The appendices for this paper are available at the following URL: http://ssrn.com/abstract=2675298

While convertible bond models recently have come to rest on solid theoretical foundation, issues in model calibration and numerical implementation still remain. This paper highlights and quantifies a number of such issues, demonstrating, among other things, that naïve calibration approaches can lead to highly significant pricing biases. We suggest a number of techniques to resolve such biases. In particular, we demonstrate how applications of the Fokker-Planck PDE allows for efficient joint calibration to debt and option markets, and also discuss volatility smile effects and the derivation of forward PDEs to embed such information into model calibration. Throughout, we rely on modern finite difference techniques, rather than the binomial or trinomial trees that so far have dominated much of the literature.

Economists have suggested a whole range of variables that predict the equity premium: dividend price ratios, dividend yields, earnings-price ratios, dividend payout ratios, corporate or net issuing ratios, book-market ratios, beta premia, interest rates (in various guises), and consumption-based macroeconomic ratios (cay). Our paper comprehensively reexamines the performance of these variables, both in-sample and out-of-sample, as of 2005. We find that [a] over the last 30 years, the prediction models have failed both in-sample and out-of-sample; [b] the models are unstable, in that their out-of-sample predictions have performed unexpectedly poorly; [c] the models would not have helped an investor with access only to information available at the time to time the market.

This paper evaluates numerous diversification strategies as a possible remedy against widespread costly investment mistakes of individual investors. Our results reveal that a very broad range of simple heuristic allocation schemes offers similar diversification gains, as well-established or recently developed portfolio optimization approaches. This holds true for both international diversification in the stock market and diversification over different asset classes. We thus suggest easy-to-implement allocation guidelines for individual investors.

We address three research questions motivated by the recent ascent of International Financial Reporting Standards (IFRS) in Europe. First, analyzing the determinants of voluntary IFRS adoption by publicly traded German firms during the period 1998-2004, we find that size, international exposure, dispersion of ownership, and recent IPOs are important drivers. Second, using the results from this determinant model to construct propensity score-matched samples of IFRS and German-GAAP (HGB) firms, we document significant differences in terms of earnings quality: IFRS firms have more persistent, less predictable and more conditionally conservative earnings. Third, analyzing information asymmetry differences between IFRS and HGB firms, we show that IFRS adopters experience a decline in bid-ask spread of 70 base points and an average of 17 more days with price changes per year. On the other hand, IFRS adopter's stock prices seem to be more volatile. In the light of some important limitations of our study, we discuss IFRS-related research opportunities in post-2005 Europe.

We measure the loss potential of Hedge Funds by combining three market risk measures: VaR, Draw-Down and Time Under-The-Water. Calculations are carried out considering three different frameworks regarding Hedge Fund returns: i) Normality and time-independence, ii) Non-normality and time-independence and iii) Non-normality and time-dependence. In the case of Hedge Funds, our results clearly state that market risk may be substantially underestimated by those models which assume Normality or, even considering Non-Normality, neglect to model time-dependence. Moreover, VaR is an incomplete measure of market risk whenever the Normality assumption does not hold. In this case, VaR results must be compared with Draw-Down and Time Under-The-Water measures in order to accurately assess about Hedge Funds loss potential.

Abstract While a routinely rebalanced portfolio such as a 60-40 equity-bond mix is commonly employed by many investors, most do not understand that the rebalancing strategy adds risk. Rebalancing is similar to starting with a buy and hold portfolio and adding a short straddle (selling both a call and a put option) on the relative value of the portfolio assets. The option-like payoff to rebalancing induces negative convexity by magnifying drawdowns when there are pronounced divergences in asset returns. The expected return from rebalancing is compensation for this extra risk. We show how a higher-frequency momentum overlay can reduce the risks induced by rebalancing by improving the timing of the rebalance. This smart rebalancing, which incorporates a momentum overlay, shows relatively stable portfolio weights and reduced drawdowns.