A Bayesian approach to bandwidth selection for multivariate kernel regression with an application to state-price density estimation

This paper presents a Bayesian approach to bandwidth selection for multivariate kernel regression. A Monte Carlo study shows that under the average squared error criterion, the Bayesian bandwidth selector is comparable to the cross-validation method and clearly outperforms the bootstrapping and rule-of-thumb bandwidth selectors. The Bayesian bandwidth selector is applied to a multivariate kernel regression model that is often used to estimate the state-price density of Arrow–Debreu securities with the S&P 500 index options data and the DAX index options data. The proposed Bayesian bandwidth selector represents a data-driven solution to the problem of choosing bandwidths for the multivariate kernel regression involved in the nonparametric estimation of the state-price density pioneered by Aït-Sahalia and Lo [Aït-Sahalia, Y., Lo, A.W., 1998. Nonparametric estimation of state-price densities implicit in financial asset prices. The Journal of Finance, 53, 499, 547.]     

Robustness of stable volatility strategies

The paper analyzes the robustness of stable volatility strategies, i.e. strategies in which the portfolio weight of the stock is inversely proportional to its local volatility. These strategies are optimal for a CRRA investor if the stock follows a diffusion process, the expected excess return is proportional to its volatility, and the hedging demand is zero. We assess the performance of stable volatility strategies when these restrictive assumptions do not hold, in particular, when the risk premium is not proportional to volatility and when the stock price is subject to jumps. We find that stable volatility strategies are indeed robust or close to robust under a maxmin decision rule. In addition to our theoretical results, we perform a simulation analysis to evaluate strategies that scale the portfolio weight by the volatility, variance or a constant portfolio weight, and also analyze the strategies using empirical excess returns. Both analyses confirm the robustness of stable volatility strategies.     

Indicator selection and stock return predictability

We propose a momentum-determined indicator-switching (MDIS) strategy, simple and effective, to improve the predictability of stock returns, which can effectively select predictors. Empirical results indicate that the stock return forecasts generated by the MDIS strategy are statistically and economically significant. And we find that super long-term momentum of predictability (SMoP) exists in predictive factors. That is, in a long period of time in the past, the best predictor among a series of factors has best prediction ability in the future. We also design restricted momentum-determined indicator-switching (RMDIS) strategy when considering economic constrain. It is robust for the prediction performance of this strategy using a series of extension and robustness test. Success of the RMDIS strategy is also seen in using technical indicators to forecast stock returns.     

Bayesian hypothesis testing in latent variable models

Hypothesis testing using Bayes factors (BFs) is known not to be well defined under the improper prior. In the context of latent variable models, an additional problem with BFs is that they are difficult to compute. In this paper, a new Bayesian method, based on the decision theory and the EM algorithm, is introduced to test a point hypothesis in latent variable models. The new statistic is a by-product of the Bayesian MCMC output and, hence, easy to compute. It is shown that the new statistic is appropriately defined under improper priors because the method employs a continuous loss function. In addition, it is easy to interpret. The method is illustrated using a one-factor asset pricing model and a stochastic volatility model with jumps.     

Mean-variance portfolio selection of cointegrated assets

This paper considers the continuous-time mean-variance portfolio selection problem in a financial market in which asset prices are cointegrated. The asset price dynamics are then postulated as the diffusion limit of the corresponding discrete-time error-correction model of cointegrated time series. The problem is completely solved in the sense that solutions of the continuous-time portfolio policy and the efficient frontier are obtained as explicit and closed-form formulas. The analytical results are applied to pairs trading using cointegration techniques. Numerical examples show that identifying a cointegrated pair with a high mean-reversion rate can generate significant statistical arbitrage profits once the current state of the economy sufficiently departs from the long-term equilibrium. We propose an index to simultaneously measure the departure level of a cointegrated pair from equilibrium and the mean-reversion speed based on the mean-variance paradigm. An empirical example is given to illustrate the use of the theory in practice.     

Agricultural arbitrage and risk preferences

A structural intertemporal model of agricultural asset arbitrage equilibrium is developed and applied to agriculture in the North Central region of the US. The data are consistent with a unifying level of risk aversion. The levels of risk aversion are more plausible than previous estimates for agriculture. However, the standard arbitrage equilibrium is rejected; perhaps, this is due to the period and the shortness of the period studied.     

Spanning tests in return and stochastic discount factor mean–variance frontiers: A unifying approach

We propose new spanning tests that assess if the initial and additional assets share the economically meaningful cost and mean representing portfolios. We prove their asymptotic equivalence to existing tests under local alternatives. We also show that unlike two-step or iterated procedures, single-step methods such as continuously updated GMM yield numerically identical overidentifying restrictions test, so there is arguably a single spanning test. To prove these results, we extend optimal GMM inference to deal with singularities in the long run second moment matrix of the influence functions. Finally, we test for spanning using size and book-to-market sorted US stock portfolios.     

Non‐linear Predictability of UK Stock Market Returns*

Linear predictability of stock market returns has been widely reported. However, recently developed theoretical research has suggested that due to the interaction of noise and arbitrage traders, stock returns are inherently non‐linear, whereby market dynamics differ between small and large returns. This paper examines whether an exponential smooth transition threshold model, which is capable of capturing this non‐linear behaviour, can provide a better characterization of UK stock market returns than either a linear model or an alternate non‐linear model. The results of both in‐sample and out‐of‐sample specification tests support the exponential smooth transition threshold model and hence the belief that investor behaviour does differ between large and small returns.     

Maximum likelihood estimation of partially observed diffusion models

This paper develops a maximum likelihood (ML) method to estimate partially observed diffusion models based on data sampled at discrete times. The method combines two techniques recently proposed in the literature in two separate steps. In the first step, the closed form approach of Aït-Sahalia (2008) is used to obtain a highly accurate approximation to the joint transition probability density of the latent and the observed states. In the second step, the efficient importance sampling technique of Richard and Zhang (2007) is used to integrate out the latent states, thereby yielding the likelihood function. Using both simulated and real data, we show that the proposed ML method works better than alternative methods. The new method does not require the underlying diffusion to have an affine structure and does not involve infill simulations. Therefore, the method has a wide range of applicability and its computational cost is moderate.     

Fund selection in target date funds

This study investigates the family-fund strategy in target date funds (TDFs). We find that fund families often create TDFs by bundling existing mutual funds and use TDFs to increase non-performance-related flows to constituent funds that are not competitive. Such incremental non-performance-related flows exceed 1% of total net assets on a monthly basis. We further investigate the potential monitoring effect of plan sponsors by examining the association between the characteristics of constituent funds and their likelihood of staying in the TDFs. We find that constituent funds with lower return performance or younger age exhibit a lower survival probability of continuing as part of a TDF. Therefore, the problem of potential agency cost at the selection point of the constituent funds may be mitigated to some extent after entering TDFs.     

Inference with non-Gaussian Ornstein–Uhlenbeck processes for stochastic volatility

Continuous-time stochastic volatility models are becoming an increasingly popular way to describe moderate and high-frequency financial data. Barndorff-Nielsen and Shephard (2001a) proposed a class of models where the volatility behaves according to an Ornstein–Uhlenbeck (OU) process, driven by a positive Lévy process without Gaussian component. These models introduce discontinuities, or jumps, into the volatility process. They also consider superpositions of such processes and we extend that to the inclusion of a jump component in the returns. In addition, we allow for leverage effects and we introduce separate risk pricing for the volatility components. We design and implement practically relevant inference methods for such models, within the Bayesian paradigm. The algorithm is based on Markov chain Monte Carlo (MCMC) methods and we use a series representation of Lévy processes. MCMC methods for such models are complicated by the fact that parameter changes will often induce a change in the distribution of the representation of the process and the associated problem of overconditioning. We avoid this problem by dependent thinning methods. An application to stock price data shows the models perform very well, even in the face of data with rapid changes, especially if a superposition of processes with different risk premiums and a leverage effect is used.     

A semiparametric stochastic volatility model

In this paper the correlation structure in the classical leverage stochastic volatility (SV) model is generalized based on a linear spline. In the new model the correlation between the return and volatility innovations is time varying and depends nonparametrically on the type of news arrived to the market. Theoretical properties of the proposed model are examined. The model estimation and comparison are conducted by Bayesian methods. The performance of the estimates are examined in simulations. The new model is fitted to daily and weekly US data and compared with the classical SV and GARCH models in terms of their in-sample and out-of-sample performances. Empirical results suggest evidence in favor of the proposed model. In particular, the new model finds strong evidence of time varying leverage effect in individual stocks when the classical model fails to identify the leverage effect.     

Spurious regressions in technical trading

This paper investigates the spurious effect in forecasting asset returns when signals from technical trading rules are used as predictors. Against economic intuition, the simulation result shows that, even if past information has no predictive power, buy or sell signals based on the difference between the short-period and long-period moving averages of past asset prices can be statistically significant when the forecast horizon is relatively long. The theoretical analysis reveals that both ‘momentum’ and ‘contrarian’ strategies can be falsely supported, while the probability of obtaining each result depends on the type of the test statistics employed.     

Testing the parametric form of the volatility in continuous time diffusion models—a stochastic process approach

We present new tests for the form of the volatility function which are based on stochastic processes of the integrated volatility. We prove weak convergence of these processes to centered processes whose conditional distributions are Gaussian. In the case of testing for a constant volatility the limiting process are standard Brownian bridges. As a consequence an asymptotic distribution free test and bootstrap tests (for testing of a general parametric form) can easily be implemented. It is demonstrated that the new tests are more than the currently available procedures. The new approach is also demonstrated by means of a simulation study.     

Local GMM estimation of time series models with conditional moment restrictions

This paper investigates statistical properties of the local generalized method of moments (LGMM) estimator for some time series models defined by conditional moment restrictions. First, we consider Markov processes with possible conditional heteroskedasticity of unknown forms and establish the consistency, asymptotic normality, and semi-parametric efficiency of the LGMM estimator. Second, we undertake a higher-order asymptotic expansion and demonstrate that the LGMM estimator possesses some appealing bias reduction properties for positively autocorrelated processes. Our analysis of the asymptotic expansion of the LGMM estimator reveals an interesting contrast with the OLS estimator that helps to shed light on the nature of the bias correction performed by the LGMM estimator. The practical importance of these findings is evaluated in terms of a bond and option pricing exercise based on a diffusion model for spot interest rate.     

A Gaussian approximation scheme for computation of option prices in stochastic volatility models

We consider European options on a price process that follows the log-linear stochastic volatility model. Two stochastic integrals in the option pricing formula are costly to compute. We derive a central limit theorem to approximate them. At parameter settings appropriate to foreign exchange data our formulas improve computation speed by a factor of 1000 over brute force Monte Carlo making MCMC statistical methods practicable. We provide estimates of model parameters from daily data on the Swiss Franc to Euro and Japanese Yen to Euro over the period 1999–2002.     

The rise of passive investing and index-linked comovement

I introduce a general equilibrium model with active investors and indexers. Indexing causes market segmentation, and the degree of segmentation is a function of the relative wealth of indexers in the economy. Shocks to this relative wealth induce correlated shocks to discount rates of index stocks. The wealthier indexers are, the greater the resulting comovement is. I confirm empirically that S&P 500 stocks comove more with other index stocks and less with non-index stocks, and that changes in passive holdings of S&P 500 stocks predict changes in comovement of index stocks.     

A model of information diffusion with asymmetry and confidence effects in financial markets

We present a model of multi-period continuous information diffusion in financial markets. We show that price and trading volume exhibit asymmetric term structures to information flow, where the diffusion rate accelerates more slowly at short horizons than it decelerates at long horizons. Bounded rationality is modelled by an endogenous trader confidence index which declines as stock price information becomes noisier, where lower confidence translates into lower trading volume and slower price accretion. Information diffusion slows and asymmetries are accentuated as traders lose confidence in information accuracy. Our empirical findings support the model's predictions of asymmetric momentum patterns and confidence effects.     

Extreme-quantile tracking for financial time series

Time series of financial asset values exhibit well-known statistical features such as heavy tails and volatility clustering. We propose a nonparametric extension of the classical Peaks-Over-Threshold method from extreme value theory to fit the time varying volatility in situations where the stationarity assumption may be violated by erratic changes of regime, say. As a result, we provide a method for estimating conditional risk measures applicable to both stationary and nonstationary series. A backtesting study for the UBS share price over the subprime crisis exemplifies our approach.     

The influence and predictive powers of mixed-frequency individual stock sentiment on stock returns

In this article, we construct mixed-frequency individual stock sentiment using MIDAS model. We first investigate the influence power of mixed-frequency individual stock sentiment on excess returns. The results indicate that the higher the frequency of individual stock sentiment is, the better it explains the variation of excess returns, that mixed-frequency individual stock sentiment, especially mixed high-frequency sentiment, exerts greater influence on excess returns than the same frequency one and that the mixed-frequency sentiment has a stronger explanatory power to the variation of excess returns than size factor, book-to-market factor, profitability factor and investment factor do. Then, we study the predictive content of mixed-frequency individual stock sentiment. The results show that the higher the frequency of individual stock sentiment is, the better the forecast performs. Moreover, by comparing the corresponding statistics in influence and predictive power models, we find that the influence power of mixed-frequency individual stock sentiment is more significant than its predictive power.