It is a major challenge for asset pricing models to generate a high equity premium and a low risk-free rate while imposing realistic consumption dynamics. To address this issue, our paper proposes a novel pricing channel: we allow for consumption drops that can spark an economic crisis. This new feature generates a large equity premium even if possible consumption drops are of moderate size. In turn, our model also matches the consumption data of 42 countries along several dimensions. In particular, our approach generates a realistic number of crises that have realistic durations and involve clustering of moderate consumption drops.

The optimal portfolio as well as the utility from trading stocks and derivatives depends on the risk factors and on their market prices of risk. We analyze this dependence for a CRRA investor in models with stochastic volatility, jumps in the stock price, and jumps in volatility. We find that the compartment of the total variance into diffusion risk and jump risk has a small impact on the utility in an incomplete market only. In contrast, the decomposition of the equity risk premium into a diffusion component and a jump risk component and the compartment of the latter into its various elements has a huge impact on the utility in a complete market. The more extreme the market prices of risk, i.e. the more they deviate from their equilibrium values, the larger the utility of the investor. Additionally, we show that the structure of the optimal exposures to jump risk crucially depends on which elements of jump risk are priced.

The paper introduces a model for the joint dynamics of asset prices which can capture both a stochastic correlation between stock returns as well as between stock returns and volatilities (stochastic leverage). By relying on two factors for stochastic volatility, the model allows for stochastic leverage and is thus able to explain time-varying slopes of the smiles. The use of Wishart processes for the covariance matrix of returns enables the model to also capture stochastic correlations between the assets. Our model offers an integrated pricing approach for both Quanto and plain-vanilla options on the stock as well as the foreign exchange rate. We derive semi-closed form solutions for option prices and analyze the impact of state variables. Quanto options offer a significant exposure to the stochastic covariance between stock prices and exchange rates. In contrast to standard models, the smile of stock options, the smile of currency options, and the price differences between Quanto options and plain-vanilla options can change independently of each other.

This paper analyzes the optimal portfolio decision of a CRRA investor in models with stochastic volatility and stochastic jumps. The investor follows a buy-and-hold strategy in the stock, the money market account, and one additional derivative. We show that both the type of the model and the structure of the risk premia have a significant impact on the optimal portfolio, on the utility gain from having access to derivatives, and on whether the investor prefers to trade OTM or ATM options. We also show that model mis-specification results in significant utility losses. Omitting jumps in volatility can be devastating, in particular if the investor chooses the seemingly optimal OTM put options. A misestimation of the structure of the risk premia has a less devastating effect, but can still lead to a loss of around 4% in the annual certainty equivalent return.

Empirical evidence shows that the implied volatility smiles for index options are significantly steeper than those for individual options. We propose a model setup where we start from the joint dynamics of the stocks and where the index value is a weighted sum of individual stock prices. Then the differences between the index smile and the smiles for individual stocks are entirely determined by the dependence structure among the stocks. We illustrate our idea in a jump-diffusion framework where both the diffusion and the jumps are decomposed into common and idiosyncratic components. Empirical data for options on the German stock index DAX and on Deutsche Bank are used to show that the model can explain the stylized facts on implied volatility smiles.

Tests for the existence and the sign of the volatility risk premium are often based on expected option hedging errors. When the hedge is performed under the ideal conditions of continuous trading and correct model specification, the sign of the premium is the same as the sign of the mean hedging error for a large class of models. We show that discrete trading and model misspecification may cause the standard test to yield unreliable results. In particular, ignoring jump risk premia can lead to incorrect conclusions. We also show that delta-gamma hedges do not increase the reliability of the test.

We consider a Lucas-type exchange economy with two heterogeneous stocks (trees) and a representative investor with constant relative risk aversion. The dividend process for one stock follows a geometric Brownian motion with constant and known parameters. The expected dividend growth rate for the other tree is stochastic and in general unobservable, although there may be a signal from which the investor can learn about its current value. We find that the equilibrium quantities in our model significantly depend on the information structure and on the level of risk aversion. While an observable stochastic drift mainly makes the economy more risky, a latent expected growth rate process with learning significantly changes the equilibrium price-dividend ratios, price reactions to dividend and drift innovations, expected returns, volatilities, correlations, and differences between the stocks. These effects are the more pronounced the more risk averse the representative investor.

We analyze the portfolio planning problem of an ambiguity averse investor. The stock follows a jump-diffusion process. We find that there are pronounced differences between ambiguity aversion with respect to diffusion risk and jump risk. Ignoring ambiguity with respect to jump risk causes larger losses in an incomplete market, whereas ignoring ambiguity with respect to diffusion risk is more severe in a complete market. For a deterministic jump size we show that the loss from market incompleteness is always increasing in the level of ambiguity aversion with respect to one risk factor and decreasing in the level of ambiguity aversion with respect to the other risk factor.

We consider an asset allocation problem in a continuous-time model with stochastic volatility and jumps in both the asset price and its volatility. First, we derive the optimal portfolio for an investor with constant relative risk aversion. The demand for jump risk includes a hedging component, which is not present in models without volatility jumps. We further show that the introduction of derivative contracts can have substantial economic value. We also analyze the distribution of terminal wealth for an investor who uses the wrong model, either by ignoring volatility jumps or by falsely including such jumps, or who is subject to estimation risk. Whenever a model different from the true one is used, the terminal wealth distribution exhibits fatter tails and (in some cases) significant default risk.