Narrow Results

The classical decision problem associated with a game is whether a given player has a winning strategy, i.e. some strategy that leads almost surely to a victory, regardless of the other players' strategies. While this problem is relevant for deterministic fully observable games, for a partially observable game the requirement of winning with probability 1 is too strong. In fact, as shown in this paper, a game might be decidable for the simple criterion of almost sure victory, whereas optimal play (even in an approximate sense) is not computable.

We therefore propose another criterion, the decidability of which is equivalent to the computability of approximately optimal play. Then, we show that (i) this criterion is undecidable in the general case, even with deterministic games (no random part in the game), (ii) that it is in the jump 0', and that, even in the stochastic case, (iii) it becomes decidable if we add the requirement that the game halts almost surely whatever maybe the strategies of the players.

In this study, the problem of seed selection is investigated. This problem is mainly treated as an optimization problem, which is proved to be NP-hard. There are several heuristic approaches in the literature which mostly use algorithmic heuristics. These approaches mainly focus on the trade-off between computational complexity and accuracy. Although the accuracy of algorithmic heuristics are high, they also have high computational complexity. Furthermore, in the literature, it is generally assumed that complete information on the structure and features of a network is available, which is not the case in most of the times. For the study, a simulation model is constructed, which is capable of creating networks, performing seed selection heuristics, and simulating diffusion models. Novel metric-based seed selection heuristics that rely only on partial information are proposed and tested using the simulation model. These heuristics use local information available from nodes in the synthetically created networks. The performances of heuristics are comparatively analyzed on three different network types. The results clearly show that the performance of a heuristic depends on the structure of a network. A heuristic to be used should be selected after investigating the properties of the network at hand. More importantly, the approach of partial information provided promising results. In certain cases, selection heuristics that rely only on partial network information perform very close to similar heuristics that require complete network data.

In this paper we propose an operational interpretation of general fuzzy measures. On the basis of this interpretation, we define the concept of coherence with respect to a partial information, and propose a rule of inference similar to the natural extension ⁷. We also suggest a definition of independence for fuzzy measures.

Given partial information, this paper considers both disruption and maintenance scheduling problem for a parallel-series system with failure interactions and hidden failures. By projection on an observed history, the filtering treatment contributes to detecting an unobservable disruption time at which one of subsystems experiences failure. The model is developed by setting it in a Markovian control framework with partial information pattern in which a control process as a repair level impacts on the system availability via maintaining both the inspection and the disruption intensity at a desirable level. Since each repair and maintenance action incurs cost, the problem is to determine an optimal control process that balances the amount of maintenance and maintenance costs. The optimal control process emerges as the solution of deterministic Hamilton Jacobi equations, and a recursive scheme is used to solve them. We illustrate the procedure for the case when the failure time of components is defined as the first passage time of a Wiener process. The proposed model is illustrated through numerical examples.

The model parameters in optimal asset allocation problems are often assumed to be deterministic. This is not a realistic assumption since most parameters are not known exactly and therefore have to be estimated. We consider investment opportunities which are modeled as local geometric Brownian motions whose drift terms may be stochastic and not necessarily measurable. The drift terms of the risky assets are assumed to be affine functions of some arbitrary factors. These factors themselves may be stochastic processes. They are modeled to have a mean-reverting behavior. We consider two types of factors, namely observable and unobservable ones. The closed-form solution of the general problem is derived. The investor is assumed to have either constant relative risk aversion (CRRA) or constant absolute risk aversion (CARA). The optimal asset allocation under partial information is derived by transforming the problem into a full-information problem, where the solution is well known. The analytical result is empirically tested in a real-world application. In our case, we consider the optimal management of a balanced fund mandate. The unobservable risk factors are estimated with a Kalman filter. We compare the results of the partial-information strategy with the corresponding full-information strategy. We find that using a partial-information approach yields much better results in terms of Sharpe ratios than the full-information approach.

A new framework for asset price dynamics is introduced in which the concept of noisy information about future cash flows is used to derive the corresponding price processes. In this framework an asset is defined by its cash-flow structure. Each cash flow is modelled by a random variable that can be expressed as a function of a collection of independent random variables called market factors. With each such "X-factor" we associate a market information process, the values of which we assume are accessible to market participants. Each information process consists of a sum of two terms; one contains true information about the value of the associated market factor, and the other represents "noise". The noise term is modelled by an independent Brownian bridge that spans the interval from the present to the time at which the value of the factor is revealed. The market filtration is assumed to be that generated by the aggregate of the independent information processes. The price of an asset is given by the expectation of the discounted cash flows in the risk-neutral measure, conditional on the information provided by the market filtration. In the case where the cash flows are the dividend payments associated with equities, an explicit model is obtained for the share-price process. Dividend growth is taken into account by introducing appropriate structure on the market factors. The prices of options on dividend-paying assets are derived. Remarkably, the resulting formula for the price of a European-style call option is of the Black–Scholes–Merton type. We consider the case where the rate at which information is revealed to the market is constant, and the case where the information rate varies in time. Option pricing formulae are obtained for both cases. The information-based framework generates a natural explanation for the origin of stochastic volatility in financial markets, without the need for specifying on an ad hoc basis the dynamics of the volatility.

We consider a reduced-form credit risk model where default intensities and interest rate are functions of a not fully observable Markovian factor process, thereby introducing an information-driven default contagion effect among defaults of different issuers. We determine arbitrage-free prices of OTC products coherently with information from the financial market, in particular yields and credit spreads and this can be accomplished via a filtering approach coupled with an EM-algorithm for parameter estimation.

We study a model of a financial market in which the dividend rates of two risky assets change their initial values to other constant ones at the times at which certain unobservable external events occur. The asset price dynamics are described by geometric Brownian motions with random drift rates switching at exponential random times, that are independent of each other and the constantly correlated driving Brownian motions. We obtain closed form expressions for the rational values of European contingent claims through the filtering estimates of occurrence of the switching times and their conditional probability density derived given the filtration generated by the underlying asset price processes.

In this study, we attempt to calculate the term structure of the interest rate under partial information using a model in which the mean reversion level of the short rate changes in accordance with a regime shift in the economy. Under partial information, an investor observes the history of only the short rate and not a regime shift; hence, calculating the term structure of the interest rate is reduced to the problem of filtering the current regime from observable short rates. Therefore, we calculate it using the filtering theory that estimates a stochastic process from noisy observations, and investigate the effects of the regime shift under partial information on the market price of risk and the volatility of a bond price compared with those under full information, in which the regime is assumed to be observable. We find that, under partial information, the regime-shift risk converts into the diffusion risk. As a result, we find that both the market price of diffusion risk and the volatility of a bond price under partial information become stochastic, even though these under full information are constant.

In this paper, the option hedging problem for a Markov-modulated exponential Lévy model is examined. We use the local risk-minimization approach to study optimal hedging strategies for Europeans derivatives when the price of the underlying is given by a regime-switching Lévy model. We use a martingale representation theorem result to construct an explicit local risk minimizing strategy.

This paper considers a stochastic control problem derived from a model for pairs trading under incomplete information. We decompose an individual asset's drift into two parts: an industry drift plus some additional stochasticity. The extra stochasticity may be unobserved, which means the investor has only partial information. We solve the control problem under both full and partial informations for utility function $U(x)=x^{1-\gamma }/(1-\gamma )$, and we make comparisons. We show the existence of stable solution to the associated matrix Riccati equations in both cases for $\gamma >1$, but for $0<\gamma <1$ there remains potential for infinite value functions in finite time. Also, we quantify the expected loss in utility due to partial information, and present a numerical study to illustrate the contribution of this paper.

This paper investigates optimal trading strategies in a financial market with multidimensional stock returns, where the drift is an unobservable multivariate Ornstein–Uhlenbeck process. Information about the drift is obtained by observing stock returns and expert opinions which provide unbiased estimates on the current state of the drift.

The optimal trading strategy of investors maximizing expected logarithmic utility of terminal wealth depends on the filter which is the conditional expectation of the drift given the available information. We state filtering equations to describe its dynamics for different information settings. At information dates, the expert opinions lead to an update of the filter which causes a decrease in the conditional covariance matrix. We investigate properties of these conditional covariance matrices. First, we consider the asymptotic behavior of the covariance matrices for an increasing number of expert opinions on a finite time horizon. Second, we state conditions for convergence in infinite time with regularly-arriving expert opinions.

Finally, we derive the optimal trading strategy of an investor. The optimal expected logarithmic utility of terminal wealth, the value function, is a functional of the conditional covariance matrices. Hence, our analysis of the covariance matrices allows us to deduce properties of the value function.

In this paper, a mean-square minimization problem under terminal wealth constraint with partial observations is studied. The problem is naturally connected to the mean–variance hedging (MVH) problem under incomplete information. A new approach to solving this problem is proposed. The paper provides a solution when the underlying pricing process is a square-integrable semi-martingale. The proposed method for study is based on the martingale representation. In special cases, the Clark–Ocone representation can be used to obtain explicit solutions. The results and the method are illustrated and supported by examples with two correlated geometric Brownian motions.

We solve the problems of mean–variance hedging (MVH) and mean–variance portfolio selection (MVPS) under restricted information. We work in a setting where the underlying price process $S$ is a semimartingale, but not adapted to the filtration $𝔾$ which models the information available for constructing trading strategies. We choose as $𝔾={𝔽}^{\mathrm{\text{det}}}$ the zero-information filtration and assume that $S$ is a time-dependent affine transformation of a square-integrable martingale. This class of processes includes in particular arithmetic and exponential Lévy models with suitable integrability. We give explicit solutions to the MVH and MVPS problems in this setting, and we show for the Lévy case how they can be expressed in terms of the Lévy triplet. Explicit formulas are obtained for hedging European call options in the Bachelier and Black–Scholes models.

In this work, we study a dynamic portfolio optimization problem related to pairs trading, which is an investment strategy that matches a long position in one security with a short position in another security with similar characteristics. The relationship between pairs, called a spread, is modeled by a Gaussian mean-reverting process whose drift rate is modulated by an unobservable continuous-time, finite-state Markov chain. Using the classical stochastic filtering theory, we reduce this problem with partial information to an equivalent one with full information and solve it for the logarithmic utility function, where the terminal wealth is penalized by the riskiness of the portfolio according to the realized volatility of the wealth process. We characterize optimal dollar-neutral strategies as well as optimal value functions under full and partial information and show that the certainty equivalence principle holds for the optimal portfolio strategy. Finally, we provide a numerical analysis for a toy example with a two-state Markov chain.

We study a credit risk model for a financial market in which the local drift rate of the logarithm of the intensity of the default time changes at the times at which certain unobservable external events occur. The risk-neutral dynamics of the default intensity are described by a generalized geometric Brownian motion and the changes of the local drift rate arrive at independent exponential times. We obtain closed form expressions for the rational values of defaultable European-style contingent claims through the filtering estimates of the occurrence of switching times given the filtration generated by the default intensity process.

We provide Galtchouk–Kunita–Watanabe representation results in the case where there are restrictions on the available information. This allows one to prove the existence and uniqueness of solution for special equations driven by a general square integrable càdlàg martingale under partial information. Furthermore, we discuss an application to risk-minimization where we extend the results of Föllmer and Sondermann, Hedging of non-redundant contingent claims, to the partial information framework and we show how our result fits in the approach of Schweizer, Risk-minimizing hedging strategies under restricted information.

In this paper, we put forward two novel schemes for probabilistic remote preparation of an arbitrary quantum state with the aid of appropriate local unitary operations when the sender and the receiver only have partial information of non-maximally entangled state, respectively. The concrete implementation procedures of the novel proposals are given in detail. Additionally, the physical realizations of our proposals are discussed based on the linear optics. Because of that neither the sender nor the receiver need to know fully the information of the partially entangled state, our schemes are useful to not only expand the application range of quantum entanglement, but also enlarge the research field of probabilistic remote state preparation (RSP).

A new framework for asset price dynamics is introduced in which the concept of noisy information about future cash flows is used to derive the corresponding price processes. In this framework an asset is defined by its cash-flow structure. Each cash flow is modelled by a random variable that can be expressed as a function of a collection of independent random variables called market factors. With each such “X-factor” we associate a market information process, the values of which we assume are accessible to market participants. Each information process consists of a sum of two terms; one contains true information about the value of the associated market factor, and the other represents “noise”. The noise term is modelled by an independent Brownian bridge that spans the interval from the present to the time at which the value of the factor is revealed. The market filtration is assumed to be that generated by the aggregate of the independent information processes. The price of an asset is given by the expectation of the discounted cash flows in the riskneutral measure, conditional on the information provided by the market filtration. In the case where the cash flows are the dividend payments associated with equities, an explicit model is obtained for the share-price process. Dividend growth is taken into account by introducing appropriate structure on the market factors. The prices of options on dividend-paying assets are derived. Remarkably, the resulting formula for the price of a European-style call option is of the Black-Scholes-Merton type. We consider the case where the rate at which information is revealed to the market is constant, and the case where the information rate varies in time. Option pricing formulae are obtained for both cases. The information-based framework generates a natural explanation for the origin of stochastic volatility in financial markets, without the need for specifying on an ad hoc basis the dynamics of the volatility.

Consider a partially informed trader who does not observe the true drift of a financial asset. Under Gaussian price dynamics with stochastic unobserved drift, including cases of mean-reversion and momentum dynamics, we take a filtering approach to solve explicitly for trading strategies which maximize expected logarithmic, exponential and power utility. We prove that the optimal strategies depend on current price and an exponentially weighted moving average (EMA) price, and in some cases current wealth, not on any other stochastic variables. We establish optimality over all price-history-dependent strategies satisfying integrability criteria, not just EMA-type strategies. Thus the condition that the optimal trading strategy reduces to a function of EMA and current price is not an assumption but rather a consequence of our analysis. We solve explicitly for the optimal parameters of the EMA-type strategies and verify optimality rigorously.