A new behavioral finance mean variance framework

PurposeThe author proposes an update to the mean variance (MV) framework that replaces a constant risk aversion parameter using a dynamic risk aversion indicator. The contribution to the literature is made through making the static risk aversion parameter operational using an indicator of market sentiment. Results suggest that Sharpe ratios improve when the author replaces the traditional risk aversion parameter with a dynamic sentiment indicator from the behavioral finance literature when allocating between a risky portfolio and a risk-free asset. However, results are mixed when using the behavioral framework to allocate between two risky assets.Design/methodology/approachThe author includes a dynamic risk aversion parameter in the mean variance framework and back test using the traditional and updated behavioral mean variance (BMV) framework to see which framework leads to better performance.FindingsThe author finds that the behavioral framework provides superior performance when allocating between a risky and risk-free asset; however, it under performs when allocating between risky assets.Research limitations/implicationsThe research is based on back testing; therefore, it cannot be concluded that this strategy will perform well in real-time circumstances.Practical implicationsPortfolio managers may use this strategy to optimize the allocation between a risky portfolio and a risk-free asset.Social implicationsAn improved allocation between risk-free and risky assets that could lead to less leverage in the market.Originality/valueThe study is the first to use such a sentiment indicator in the traditional MV framework and show the math.

Market expectation shifts in option-implied volatilities in the US and UK stock markets during the Brexit vote

This paper investigates the effect of the Brexit vote on the connection between UK stock market expectations and US stock market returns. To gauge UK stock market expectations, the option-implied volatilities of the FTSE 100 index are calculated in the period starting five months before and ending four months after the Brexit referendum. To keep the analysis “clean”, it stops right before the 2016 US presidential elections. It uses an OLS regression to estimate the change in the relationship between US and UK stock market expectations.The main findings show that the US and UK stock markets became somewhat less integrated four months after the Brexit referendum compared to the five months before it. The S&P 500 Index returns have a statistically significant impact on implied volatilities of the FTSE 100 only before the Brexit referendum. However, the British risk-free rate (LIBOR) became a statistically significant factor affecting FTSE 100 implied volatilities only after Brexit. This analysis may be used by decision-makers in the money management industry to act appropriately during Black Swan events. When UK citizens unexpectedly voted in favor of Brexit, the risk-free rate dropped, making it cheaper to invest, increasing the Sharpe ratios of equity portfolios. Coupled with increased uncertainty, this caused portfolio reallocations. In turn, expected volatility measured by options-implied volatility increased. AcknowledgmentThe authors would like to thank Olesia Verchenko for critique, a KSE M.A., external defense reviewer for helpful comments.

Daily Cryptocurrency Returns Forecasting and Trading via Machine Learning

We execute a comparative analysis of machine learning models for the time-series forecasting of the sign of next-day cryptocurrency returns. We begin by compiling a proprietary dataset that encompasses a wide array of potential cryptocurrency valuation factors (price trends, liquidity, volatility, network, production, investor attention), subsequently identifying and evaluating the most significant factors. We apply eight machine learning models to the dataset, utilizing them as classifiers to predict the sign of next day price returns for the three largest cryptocurrencies by market capitalization: bitcoin, ethereum, and ripple. We show that the most significant valuation factors for cryptocurrency returns are price trend variables, seven and thirty-day reversal, to be specific. We conclude that support vector machines result in the most accurate classifications for all three cryptocurrencies. Additionally, we find that boosted models like AdaBoost and XGBoost have the poorest classification accuracy. At length, we construct a probability-based trading strategy that secures either a daily long or short position on one of the three examined cryptocurrencies. Ultimately, the strategy yields a Sharpe of 2.8 and a cumulative log return of 3.72. On average, the strategy’s log returns outperformed standalone investments in all three cryptocurrencies by a factor of 5.64, and Sharpe ratios more than threefold.

CCMV Portfolio Optimization with Stocks and Options using Forecasted Data

Although the future of a financial market is ambiguous and mysterious, historical data play a key role to forecast the future of the market. Along with all the advantages of these data, they may result to some errors and consequently, some losses. In this paper, we consider the cardinality constraints mean-variance (CCMV) portfolio optimization model in the presence of short selling, risk-neutral interest rate and transaction costs. We insure the investment using options against unfavorable outcomes. The Geometric Brownian Motion model is utilized to forecast the stocks prices. Also, to improve the results, we calibrate its parameters using historical data by the maximum likelihood estimation method. We perform numerical experiments using historical and forecasted data on the S&P 500 index, to assess the efficiency of the GBM model in forecasting stocks prices. Also, to examine the effect of options in the portfolio, we compare the portfolio with stocks only versus the portfolio with stocks and options using historical and forecasted data in terms of returns and Sharpe ratios.

EQUITY MARKETS RISKS AND RETURNS: IMPLICATIONS FOR GLOBAL PORTFOLIO CAPITAL FLOWS DURING PANDEMIC AND CRISIS PERIODS

The paper is devoted to the study of risk and return tradeoff in the global equity market as well as particular market groups: developed, emerging and frontier markets. Impact of this tradeoff on international equity portfolio liabilities is explored. The study confirms the hypothesis that there are some specific patterns of risk and return tradeoff during crisis periods and periods of markets regular regime that substantially differ from each other and define global portfolio equity flows and liabilities in a specific way. The paper thus carries out its main objective that implies revealing these patterns with respective qualitative features and quantitative markers, specifying their implications for equity portfolio flows to markets of different types. Risks and returns for different market groups and global market as a whole are calculated for the period between 2002 and 2020 using standard methodology of contemporary portfolio theory and MSCI indices monthly values. The data for international equity portfolio liabilities as well as the share of particular market group in the global market are used as dependent variables. The latter are regressed by calculated risks and returns. Using the model results and some analytical developments, two patterns of risk/return tradeoff are discovered. The pattern attributable to regular market regimes is characterized by positive returns which is 1.51 % in average for the global market, 1.48 % for developed markets and 2.03 % for emerging markets. Risks in regular pattern are relatively small or moderate at the average level of 3.05 for the global market and are all below the median (3.48). Respective risks for developed and emerging markets are 3.02 and 4.54. The Sharpe ratios in regular pattern are positive at the average level of 0.60 for the global market, 0.57 and 0.45 for developed and emerging market groups respectively. The crisis pattern implies negative returns at the mean of -1.04 for the global market, -0.97 for the developed group and -1.35 for the emerging markets. High risks are all above the median and in average compile 5.5 for the global market, 5.47 for the developed markets and 6.68 for the emerging group. Sharpe ratios for this pattern are negative being equal to -0.19 in the mean. The average value is -0.18 for developed markets and it is -0.24 for emerging markets. Specific pattern of 2020 crisis should be settled out. Its main feature that substantially distinguishes it from other crises is the combination of highest risk level and the positive returns at the same time. Elaborated regression model confirms the direct impact of return and indirect impact of risk on global portfolio liabilities. The influence of risk for regular and crisis patterns does not differ substantially while the impact of return is much stronger during periods of increased volatility (respective model parameters are 3793.76 and 447.24). However, the discovered impact is much more reliable in crisis pattern that is supported by much higher determination ratio. Developed markets experience similar effects.

Equity market anomalies in major European economies

This paper investigates five leading equity market anomalies – size, value, momentum, profitability, and asset growth, for four Western European markets, namely, Germany, France, Italy and Spain, from January 2002 to March 2018. The study tests whether these anomalies reverse under different macro-economic uncertainty conditions, and evaluates if strategies based on time diversification can be formed using these equity market anomalies. Market anomalies were tested using four major asset pricing models – the Capital Asset Pricing Model, the Fama-French three-factor model, the Carhart model, and the Fama-French five-factor model. Macro-economic uncertainty was tested using two proxies, namely VIX and default premiums. Time diversified strategies were examined by estimating Sharpe ratios of combined portfolios formed by combining winner univariate portfolios. Value effect in Germany, Size effect in France and Profitability effect in Italy and Spain provide the highest unadjusted returns on long side strategies. No significant reversal of these anomalies was found under different macroeconomic uncertainties. Asset pricing tests show that CAPM works well for Spain and Italy, while Carhart’s model explains returns in Germany. The Fama-French five factor model does not seem to be a good descriptor of asset pricing for data. No suitable model for explaining asset returns is identified for France. Finally, it is observed that some of the equity market anomalies seem to be countercyclical and therefore provide time diversification opportunities. The study has implications for academicians, investors, and policy makers by providing insights for developing profitable investment strategies and highlighting the efficacy of alternative models as performance benchmarks.

Portfolio management system in equity market neutral using reinforcement learning

Portfolio management involves position sizing and resource allocation. Traditional and generic portfolio strategies require forecasting of future stock prices as model inputs, which is not a trivial task since those values are difficult to obtain in the real-world applications. To overcome the above limitations and provide a better solution for portfolio management, we developed a Portfolio Management System (PMS) using reinforcement learning with two neural networks (CNN and RNN). A novel reward function involving Sharpe ratios is also proposed to evaluate the performance of the developed systems. Experimental results indicate that the PMS with the Sharpe ratio reward function exhibits outstanding performance, increasing return by 39.0% and decreasing drawdown by 13.7% on average compared to the reward function of trading return. In addition, the proposed model is more suitable for the construction of a reinforcement learning portfolio, but has 1.98 times more drawdown risk than the . Among the conducted datasets, the PMS outperforms the benchmark strategies in TW50 and traditional stocks, but is inferior to a benchmark strategy in the financial dataset. The PMS is profitable, effective, and offers lower investment risk among almost all datasets. The novel reward function involving the Sharpe ratio enhances performance, and well supports resource-allocation for empirical stock trading.

Evolution of REITs in the Nigerian real estate market

PurposeThis study aims to examine the performance of real estate investment trusts (REITs) in emerging property markets. The paper used the Nigerian REIT (N-REIT) as a case study of an African REIT market, to provide information for investment decisions.Design/methodology/approachSeven years quarterly returns data (from 2013 to 2019) were obtained and used to analyse the holding period returns, return–risk ratio, coefficient of variation and Sharpe ratios of N-REIT, All Share Index of stocks (ASI) and the Federal Government Bonds (FGB) in Nigeria.FindingsThe study reveals that N-REIT outperformed stocks but underperformed bonds. Concerning risk, stocks provided the highest level of risk (7.69), followed by bonds (2.78), while N-REIT provided the lowest risk (2.7). The Sharpe ratios showed that N-REIT is the second-best performing asset, while bond is the first and stocks the last on the risk-adjusted basis.Practical implicationsN-REIT is the second-largest REIT market in Africa with a market capitalisation of about US$136m. The N-REIT market has provided investment benefits to institutional and individual investors such as liquidity, transparency and ease of transaction. This study shows the peculiarity of N-REITs; this can guide investors in making informed investment decisions.Originality/valueThis study is one of the first to empirically analyse in a comparative context, the risk-adjusted performance of N-REITs, ASI and FGB. The study will add to the limited research in this field and equip investors with valuable information for informed investment decisions.

Fundamental Analysis and Mean-Variance Optimal Portfolios

We integrate fundamental analysis with mean-variance portfolio optimization to form fully optimized fundamental portfolios. We find that fully optimized fundamental portfolios produce large out-of-sample factor alphas with high Sharpe ratios. They substantially outperform equal-weighted and value-weighted portfolios of stocks in the extreme decile of expected returns, an approach commonly used in fundamental analysis research. They also outperform the factor-based and parametric portfolio policy approaches used in the prior portfolio optimization literature. The relative performance gains from mean-variance optimized fundamental portfolios are persistent through time, robust to eliminating small capitalization firms from the investment set, and robust to incorporating estimated transactions costs. Our results suggest that future fundamental analysis research could implement this portfolio optimization approach to provide greater investment insights.