Mean-variance Investor Research Articles

Forecasting Realized Volatility: The Choice of Window Size

ABSTRACTDifferent window sizes may produce different empirical results. However, how to choose an ideal window size is still an open question. We investigate how the window size affects the predictive performance of volatility. The empirical results show that the loss function for volatility prediction takes on a U‐shape as the window size increases. This suggests that if the window size is chosen too large or too small, the loss function tends to be large and the model's predictive accuracy decreases. A window size of between 1000 and 2000 observations is ideal for various assets because it can produce relatively minimal forecast errors. From an asset allocation perspective, a mean–variance investor can obtain sizeable utility by using a model with a low loss function value for her portfolio. Moreover, the results are robust in a variety of settings.

Forecasting Chinese Stock Market Volatility With Volatilities in Bond Markets

ABSTRACTIn this paper, we investigate whether the bond markets contain important information that can improve the accuracy of stock market volatility forecasts in China. We use realized volatility (RV) implemented by different maturity treasury bond futures contracts to predict the Chinese stock market volatility. Our work is based on the heterogeneous autoregressive (HAR) framework. Empirical results show that the volatility of treasury bond contracts with longer maturities (especially 10 years) has the best effect on predicting the Chinese stock market volatility, both in sample and out of sample. Two machine learning methods, the scaled principal component analysis (SPCA) and the least absolute shrinkage and selection operator (lasso), are also more effective than the HAR benchmark model's prediction. Finally, mean–variance investors can achieve substantial economic gains by allocating their investment portfolios based on volatility forecasts after introducing treasury bond futures volatility.

Forecasting crude oil price: A deep forest ensemble approach

We made an application of a cutting-edge machine learning method, the deep forest ensemble approach (DFEA, Zhou and Feng (2017, 2019)), to empirically predict crude oil prices. We used a large data set with 36 explanatory variables to compare the predictability of the DFEA with seven popular machine learning models and their mean combination method. The out-of-sample forecasting results showed that the DFEA statistically and economically outperforms all the competing models in terms of out-of-sample R square and success ratio. The DFEA also displayed sizable certainty equivalent return (CER) gains for a mean-variance investor in practice from an asset allocation perspective. Furthermore, we found that the predictive power of the DFEA stems from technical indicators, especially momentum predictors. Our results survived in various robustness checks.

Weathering market swings: Does climate risk matter for agricultural commodity price predictability?

The challenges posed by climate change on the agricultural market have become a pressing concern. An accurate reading of future agricultural commodity prices can be an invaluable planning instrument for diverse interested parties. Here, we explore asset pricing implications of climate risk for the agricultural commodity market from January 2005 to December 2021. Through introducing a composite climate risk index based on the four individual climate risk measures of Faccini et al. (2023), our findings provide valuable insights into the time-series predictability of aggregate climate risk on future agricultural commodity returns, both in- and out-of-sample. This powerful predictability conveys substantial economic benefits to mean–variance investors and cannot be subsumed by conventional economic predictor variables. The evidence further suggests that physical risk, especially global warming, exhibits much stronger return predictability than transition risk. Moreover, we emphasize the pivotal role of climate risk in shaping supply dynamics and capturing investor attention, thereby serving as potential drivers of return predictability. Overall, these predictive insights hold important implications for risk management, investment strategies, and policy formulation in the agricultural commodity market.

Informed Trading and Return Predictability in China: Research Based on Ensemble Neural Network

ABSTRACT We construct a new informed trading index based on the high-frequency trading data of the Chinese A-share market using the ensemble neural network algorithm. We find that the informed trading index is a strong negative predictor of future aggregate stock market returns, with monthly in-sample and out-of-sample R 2 of 5.45% and 3.53%, respectively, which is far greater than the predictive power of other previously studied informed trading indicators and macroeconomic variables. The asset allocation strategy based on our index can generate large economic gains for the mean-variance investors, with annualized CER (certain equivalent return) gains ranging from 10.91% to 7.80% as the investor’s risk appetite varies. The driving force of the predictive power appears to stem from the liquidity provider role that informed traders play, which decreases the market’s illiquidity risk and lowers the risk premium of equity. Our analysis complements the returns predictability study by adding a new predictor on the one hand and informs market timing strategies on the other.

Biodiversity and stock returns

This study constructs the Biodiversity Risk (BR) index to investigate the impact of biodiversity risk challenges on financial markets. In this study, biodiversity risk is defined as the concern for risks to biodiversity and the associated negative impacts. The findings indicate that, both in- and out-of-sample, the BR index demonstrates robust predictive ability for the market risk premium. Moreover, the BR index contributes significantly to the asset allocation of mean-variance investors, yielding notable economic benefits. This study reveals industry heterogeneity in the impact of biodiversity on stock returns. Furthermore, the index demonstrates stronger predictive power for stocks in regions with larger nature reserves, non-state-owned enterprises, and high GDP. Additionally, this study quantifies biodiversity physical risk and transition risk for the first time, showing that physical risk significantly affects stock returns, but the impact of transition risk has yet to materialize.

Predicting the equity premium with financial ratios: A comprehensive look over a long period in Korea

This study investigates whether financial ratios, such as the dividend-price, price-to-book, or price-earnings ratios, reliably predict the equity premium in Korea. We construct the longest possible time series data for the stock market index excess returns including dividends and associated financial ratios by collecting data from various sources including data only available in hard copy publications from 1968 to 2021. We also construct the cyclically adjusted price-earnings (CAPE) ratio as a predictor which has not been used in previous studies on the Korean stock market. We estimate the predictive regressions using the IVX method, which is robust to having a persistent independent variable and overlapping data as a dependent variable and can accommodate multiple independent variables. We also assess the economic gains of the predictability by examining the certainty equivalent return for a mean-variance investor. We find that the CAPE ratio reliably predicts the equity premium for horizons ranging from one month to two years, and the results survive various robustness tests.

Forecasting stock market returns with a lottery index: Evidence from China

AbstractThis study constructs a Chinese lottery index (LI) based on six popular lottery preference variables by using the partial least squares method and examines the relationship between the LI and future stock market returns during the period from January 2000 to December 2021. We find that the LI can negatively predict stock market excess returns in‐sample and out‐of‐sample. In addition, the LI can generate a large economic gain for a mean–variance investor. Finally, the predictive sources of the LI stem from a cash flow channel and can be explained by the positive volume–volatility relationship and investor attention.

Optimal mean-variance investment and reinsurance strategies with a general Lévy process risk model

This paper is concerned with the optimal time-consistent investment and reinsurance strategies for mean-variance insurers with a general Lévy Process model. Expressly, the insurers are allowed to purchase proportional reinsurance and invest in a financial market, where the surplus of the insurers is assumed to follow a Cramér–Lundberg model and the financial market consists of one risk-free asset and one risky asset whose price process is driven by a general Lévy process. Through the verification theorem, the closed-form expressions of the optimal strategies under the mean-variance criterion are derived by a complex partial integral differential Hamilton–Jacobi–Bellman equations. Finally, numerical simulations are provided to verify the effectiveness of the proposed optimal strategies and some economic interpretations are drawn.

Variational Inference for Large Bayesian Vector Autoregressions

We propose a novel variational Bayes approach to estimate high-dimensional Vector Autoregressive (VAR) models with hierarchical shrinkage priors. Our approach does not rely on a conventional structural representation of the parameter space for posterior inference. Instead, we elicit hierarchical shrinkage priors directly on the matrix of regression coefficients so that (a) the prior structure maps into posterior inference on the reduced-form transition matrix and (b) posterior estimates are more robust to variables permutation. An extensive simulation study provides evidence that our approach compares favorably against existing linear and nonlinear Markov chain Monte Carlo and variational Bayes methods. We investigate the statistical and economic value of the forecasts from our variational inference approach for a mean-variance investor allocating her wealth to different industry portfolios. The results show that more accurate estimates translate into substantial out-of-sample gains across hierarchical shrinkage priors and model dimensions.

Dissecting climate change risk and financial market instability: Implications for ecological risk management.

This research investigates the impact of climate challenges on financial markets by introducing an innovative approach to measure climate risk, specifically the aggregate climate change concern (ACCC) index. The study aims to assess and quantify the potential influence of climate change and risk-related factors on the performance and dynamics of financial markets. In this paper, concern is defined as the attention paid to the risk of climate change and the associated negative consequences. The findings demonstrate that the aggregate index exhibits robust predictability of market risk premiums, both within the sample and out-of-sample. By comparison, the index contains additional information beyond 14 economic predictors and 12 risk/uncertainty indexes in forecasting stock market return. In addition, the index proves valuable for mean-variance investors in asset allocation, leading to significant economic gains. The study identifies the index's ability to capture the reversal of temporary price crashes caused by overreactions to climate change risk. Furthermore, it exhibits stronger return forecasting capability for green stocks, non-state-owned enterprise (non-SOE) stocks, and stocks in regions with low air pollution. Particularly during periods of low air pollution and relaxed regulation, the index displays an enhanced ability to forecast returns. The study's findings provide valuable insights for policymakers and financial institutions as they address 21st-century environmental challenges. Moreover, these findings can inform the design of adaptive measures and interventions aimed at mitigating ecological risks and promoting sustainable economicgrowth.

Quantifying the non-Gaussian gain

In this paper, we quantify the economic gain from accounting for departures from normality for the mean-variance (MV) investor. We provide two models that account for the key empirical regularities of financial returns: stochastic volatility, asymmetric returns, heavy tails and tail dependence. We show that accounting for departures from normality leads to significant gains in expected utility commensurate with or exceeding typical active management fees. The majority of the uplift in expected utility derives from accounting for stochastic volatility.

Out‐of‐sample volatility prediction: Rolling window, expanding window, or both?

AbstractEstimation windows, either rolling or expanding, are used for volatility forecasting. In this study, we propose a new approach relying on both estimation windows. Our method is based on how well these two windows performed in terms of prediction during a recent period of past time. We will continue to use whichever one has performed better in the past. Results show that our strategy significantly outperforms the individual and mean combination models. Whether the window is rolling or expanding, the relatively better performance is persistent. In other words, we document the existence of the momentum of predictability (MoP). A mean–variance investor can achieve the highest utility gains using our strategy for volatility forecasting. Moreover, the results pass a series of robustness tests.

Industry volatility spillover and aggregate stock returns

ABSTRACT We propose an industry volatility spillover index and find that this new predictor has an impressive in-sample and out-of-sample predictability of aggregate stock returns. A one-standard-deviation increase in this index leads to a 4.14% increase in the expected excess return over the next month. Furthermore, a mean-variance investor can realize sizable economic gains by using this spillover index in his/her asset allocation. The forecasting power of the spillover index remains significant after controlling for popular economic variables and newly proposed predictors. Due to the absence of learning, the spillover index shows increasing predictive ability over the recent period. The economic origins of the spillover index’s success stem from both the cash flow and discount rate channels, as well as significant association with investor sentiment and tail risk. The return predictability of this index is pervasive across characteristic-sorted portfolios and is particularly strong for large-cap or difficult-to-value stocks.

Mean-Variance Investment Without Risk-Free Assets in PT Company Shares PT Ace Hardware (Aces.Jk), PT Mayora Indah (Myor.Jk), PT Bri (Bbri.Jk), PT Siloam Hospital (Silo.Jk), PT Eterindo Wahanatama (Etwa.Jk)

Portfolio is a form of strategy that investors often apply in risky investment conditions. The essence of portfolio construction is to allocate funds to various investment options to minimize investment risk. Therefore, the aim of this discussion is to construct an investment portfolio of several shares using an average variable portfolio optimization model without risk-free assets. To obtain an optimal portfolio, a mean-variance investment optimization model without risk-free assets or what is called the Basic Markowitz model is used. This involves investors measuring the risk of an asset using its “variance” and then comparing it to the asset's average. It is hoped that this discussion can help investors to obtain an optimal portfolio, especially from the five selected shares.

Maximum principle for conditional mean-field FBSDEs systems with regime-switching involving impulse controls

In this work, a stochastic control problem for a forward-backward regime-switching system with mean-field interactions involving impulse controls is studied. One feature of this study lies in the conditional mean-field term used in the model. Necessary conditions for both regular and impulse controls are established via convex variational technique, and the verification theorems are obtained under appropriate convexity assumptions as well. A mean-variance investment and consumption example is discussed to test the validity of the theoretical results, and numerical simulations are also carried out to make the results clearer.

Policy uncertainty and stock market volatility revisited: The predictive role of signal quality

AbstractThis paper provides novel insight into the growing literature on the policy uncertainty‐stock market volatility nexus by examining the out‐of‐sample predictive ability of the quality of political signals over stock market volatility at various forecast horizons. Specifically, we examine whether or not accounting for the signal quality in forecasting models within a mixed frequency framework can improve forecast performance and help achieve economic gains for investors. Both in‐ and out‐of‐sample tests, based on a GARCH‐MIDAS framework, show that the quality of the policy signal matters regarding the predictive role of policy uncertainty over subsequent stock market volatility. While high economic policy uncertainty (EPU) predicts high volatility, particularly when the signal quality is high, the positive relationship between EPU and volatility breaks down when the signal quality is low. The improved out‐of‐sample volatility forecasts obtained from the models that account for the quality of policy signals also help typical mean–variance investors achieve improved economic outcomes captured by higher certainty equivalent returns and Sharpe ratios. Although our results indicate clear distinctions between the US and UK stock markets in terms of how market participants process policy signals, they highlight the role of the quality of policy signals as a driver of volatility forecasts with significant economic implications.

Can climate change attention predict energy stock returns?

We propose a climate change attention (CCA) index based on Google search volume index (GSVI) from 2004 to 2021 and show that it is an economically and statistically significant negative predictor for next month's energy stock returns. The index is extracted using principal component analysis (PCA), but the results are similar by using the equal-weighted average method. Compared with 14 traditional macroeconomic predictors, CCA performs the best and provides complementary information when added into bivariate and multivariate macro predictive models. When further considering the effect of CCA's forecasting power over different periods, strong evidence is shown that this outperformance is especially prominent in economic depressions and down market conditions. From the asset allocation perspective, CCA can provide a mean-variance investor with significant economic gains under alternative risk aversions. Our empirical results prove that investors' attention to climate change contains predictive information for excess returns of global traditional energy stock index.

Forecasting stock market volatility with various geopolitical risks categories: New evidence from machine learning models

This paper investigates how geopolitical risks influence the prediction performance on the US stock market volatility with machine learning models. Further, it compares the predictive performance of individual and combination forecast methods. With SHAP algorithm, it could identify which factor has a great impact and fully extract the information of geopolitical risks in predicting. Empirical results show that military build-ups and escalation of war have great importance on predicting realized volatility among various geopolitical risks. The research further emphasizes the superior performance of machine learning and forecast combination methods, especially SVR method and trimmed mean combination. In addition, by allocating portfolio according to the machine learning-based volatility forecasts, particularly elastic net and random forest, a mean-variance investor can achieve sizeable financial benefits. Our paper provides substantial implications for political risk management and volatility forecasting.

Default return spread: A powerful predictor of crude oil price returns

AbstractThis paper uses the default return spread (DFR) to predict crude oil price returns over the period January 1986 through December 2020. Results of in‐sample and out‐of‐sample analyses show that the DFR can predict oil price returns and significantly outperform the benchmark and other competing variables. In an asset allocation exercise, a mean–variance investor can obtain considerable certainty equivalent return (CER) gains based on the return forecasts of DFR relative to the benchmark. We also perform a series of robustness tests to confirm our previous conclusion. We further investigate the source of the DFR's predictive ability from oil market sentiment, in which we provide some theoretical basis to explain.