Excessive Volatility is Also a Feature of Individual Level Forecasts

The excessive volatility of prices in financial markets is one of the most pressing puzzles in social science. It has led many to question economic theory, which attributes beneficial effects to markets in the allocation of risks and the aggregation of information. In exploring its causes, we investigated to what extent excessive volatility can be observed at the individual level. Economists claim that securities prices are forecasts of future outcomes. Here, we report on a simple experiment in which participants were rewarded to make the most accurate possible forecast of a canonical financial time series. We discovered excessive volatility in individual-level forecasts, paralleling the finding at the market level. Assuming that participants updated their beliefs based on reinforcement learning, we show that excess volatility emerged because of a combination of three factors. First, we found that submitted forecasts were noisy perturbations of participants’ revealed beliefs. Second, beliefs were updated using a prediction error based on submitted forecast rather than revealed past beliefs. Third, in updating beliefs, participants maladaptively decreased learning speed with prediction risk. Our results reveal formerly undocumented features in individual-level forecasting that may be critical to understand the inherent instability of financial markets and inform regulatory policy.     

A TEST OF A GENERAL EQUILIBRIUM STOCK OPTION PRICING MODEL

An empirical version of the Cox, Ingersoll, and Ross (1985a) call option pricing model is derived, assuming execution price uncertainty in the options market. the pricing restrictions come in the form of moment conditions in the option pricing error. These can be estimated and tested using a version of the method of simulated moments (MSM). Simulation estimates, obtained by discretely approximating the risk-neutral processes of the underlying stock price and the interest rate, are substituted for analytically unknown call prices. the asymptotics and other aspects of the MSM estimator are discussed. the model is tested on transaction prices at 15-minute intervals. It substantially outperforms the Black-Scholes model. the empirical success of the Cox-Ingersoll-Ross model implies that the continuous-time interest rate implicit in synchronous transaction quotes of 90-day Treasury-bill futures contracts is an-albeit noisy-proxy for the instantaneous volatility on common stock. the process of the instantaneous volatility is found to be close to nonstationary. It is well approximated by a heteroskedastic unit-root process. With this approximation, the Cox-Ingersoll-Ross model only slightly overprices long-maturity options.     

A general equilibrium model of changing risk premia: theory and tests

We derive and test a dynamic discrete-time model of asset returns.Both the risks of individual securities and equilibrium riskpremia change predictably in the model, but these changes canbe attributed to movements in the returns and prices of onlytwo well-diversified portfolios. Any other components of returnsshould be unpredictable. Using the generalized method of moments,the model is estimated and tested on portfolios of equities.We find the data supportive of the model's restrictions, evenwhen instruments designed to capture the January effect areemployed.     

Market microstructure effects of government intervention in the foreign exchange market

As asymmetric information model of the bid - ask spread is developedfor a foreign exchange market subject to occasional governmentinterventions. Traditional tests of the unbiasedness of theforward rate as a predictor of the future spot rate are shownto be inconsistent when the rates are measured as the averageof their respective bid and ask quotes. Larger bid - ask spreadson Fridays are documented. Reliable evidence of asymmetric bid- ask spreads for all days of the week, albeit more pronouncedon Fridays, are presented. The null hypothesis that the forwardrate is an unbiased predictor of the future spot rate continuesto be rejected. The regression slope coefficients increase towardunity, however, indicating a less variable risk premium.     

Asset Prices and Trading Volume in a Beauty Contest

Speculators buy an asset hoping to sell it later to investors with higher private valuations. If agents are uncertain about the distribution of private valuations and about the beliefs of others about this distribution, a beauty contest with an infinite hierarchy of beliefs arises. Under Harsanyi's assumption of a common prior the infinite beliefs hierarchy is readily solved using Bayes' law. This paper shows that common knowledge of the "beliefs formation rule," mapping the private valuation of each gent into his first-order belief, also simplifies the beliefs hierarchy while allowing for disagreement among agents. We analyse the resulting speculation in a stylized asset market. Several statistics, computed only from readily observable quote, return and volume data, are evaluated in terms of their power to discriminate between genuine disagreement and the Harsanyian case. Only statistics that relate volume and volatility, or volume and changes in best offers, have the necessary discriminatory power.     

Inducing liquidity in thin financial markets through combined-value trading mechanisms

Asset pricing theory hypothesizes that investors are only interested in portfolios; individual securities are evaluated only in terms of their contribution to portfolio risk and return. Yet, standard financial market design is that of parallel, unconnected markets, whereby investors cannot submit orders in one market conditional on events in others. When markets are thin, this exposes them to substantial execution risk. Fear of ending up with unbalanced portfolios after trading may even keep investors from submitting orders, further eroding liquidity and the ability of markets to equilibrate. The suggested solution is a portfolio trading mechanism referred to as combined-value trading (CVT). Investors are allowed to submit orders for packages of securities and the system matches trades and computes prices by optimally combining portfolio orders in an open book. We study the performance of the CVT mechanism experimentally and compare it to the performance of parallel, unconnected double auctions in experiments with similar parametrization and either a similar number of subjects or substantially thicker markets. We present evidence that our portfolio trading mechanism facilitates equilibration to the extent that the thicker markets do. Inspection of order submission and trade activity reveals that subjects manage to exploit the direct linkages between markets enabled by the CVT system.     

Filtering Returns for Unspecified Biases in Priors when Testing Asset Pricing Theory

Procedures are presented that allow the empiricist to estimate and test asset pricing models on limited-liability securities without the assumption that the historical payoff distribution provides a consistent estimate of the market's prior beliefs. The procedures effectively filter return data for unspecified historical biases in the market's priors. They do not involve explicit estimation of the market's priors, and hence, economize on parameters. The procedures derive from a new but simple property of Bayesian learning, namely: if the correct likelihood is used, the inverse posterior at the true parameter value forms a martingale process relative to the learner's information filtration augmented with the true parameter value. Application of this central result to tests of asset pricing models requires a deliberate selection bias. Hence, as a by-product, the article establishes that biased samples contain information with which to falsify an asset pricing model or estimate its parameters. These include samples subject to, e.g. survivorship bias or Peso problems.     

Local parametric analysis of derivatives pricing and hedging

A novel methodology for the analysis of derivatives pricing in incomplete markets is tested empirically. The methodology generates hedge ratios and derivatives prices. They are estimated from the correlation structure between the local co-movements of securities prices. First, the hedge ratios from a parsimonious complete-market model are estimated by fitting locally the changes in the derivatives and the underlying securities prices. Second, derivatives prices are obtained from the locally estimated hedge ratios. The methodology, referred to as local parametric estimation, is tested on a dataset of DAX index options and futures transactions from the computerized German Futures Exchange.     

An Exploration of Neo-Austrian Theory Applied to Financial Markets

We attempt to translate Neo-Austrian ideas about the workings of financial markets, as originally advanced by F. A. Hayek, into the standard probabilistic language of modern finance. We focus on an apparent paradox, namely the insistence of Neo-Austrians on order (i.e., stationarity) together with ever-reemerging inefficiencies . The paper's findings have implications beyond Neo-Austrian theory: They demonstrate how easy it is to reject market efficiency, but how much more difficult it is to discern the nature of the inefficiency . We illustrate our findings with price data from the U.S. Treasury bill market over the period 1962 to 1999. There is ample evidence that the price of a three-month Treasury bill is not a random walk, yet the sign of the average price change is erratic, so that inference about the nature of the inefficiency is unreliable.     

Implementing statistical criteria to select return forecasting models: what do we learn?

Statisticl model selection criteria provide an informed choiceof the model with best external (i.e., out-of-sample) validity.Therefore they guard against overfitting ('data snooping').We implement several model selection criteria in order to verifyrecent evidence of predictability in excess stock returns andto determine which variables are valuable predictors. We confirmthe presence of in-sample predictability in an internationalstock market dataset, but discover that even the best predictionmodels have no out-of-sample forecasting power. The failureto detect out-of-sample predictability is not due to lack ofpower.