Operations Research
http://lists.repec.orgmailman/listinfo/nep-ore
Operations Research
2016-11-27
Time-varying Combinations of Bayesian Dynamic Models and Equity Momentum Strategies
http://d.repec.org/n?u=RePEc:tin:wpaper:20160099&r=ore
A novel dynamic asset-allocation approach is proposed where portfolios as well as portfolio strategies are updated at every decision period based on their past performance. For modeling, a general class of models is specified that combines a dynamic factor and a vector autoregressive model and includes stochastic volatility, denoted by FAVAR-SV. Next, a Bayesian strategy combination is introduced in order to deal with a set of strategies. Our approach extends the mixture of the experts analysis by allowing the strategic weights to be dependent between strategies as well as over time and to further allow for strategy incompleteness. Our approach results in a combination of different portfolio strategies: a model-based and a residual momentum strategy. The estimation of this modeling and strategy approach can be done using an extended and modified version of the forecast combination methodology of Casarin, Grassi, Ravazzolo and Van Dijk(2016). Given the complexity of the non-linear and non-Gaussian model used a new and efficient filter is introduced based on the MitISEM approach by Hoogerheide, Opschoor and Van Dijk (2013). Using US industry portfolios between 1926M7 and 2015M6 as data, our empirical results indicate that time-varying combinations of flexible models in the FAVAR-SV class and two momentum strategies lead to better return and risk features than very simple and very complex models. Combinations of two strategies help, in particular, to reduce risk features like volatility and largest loss, which indicates that complete densities provide useful information for risk.
Nalan Basturk
Stefano Grassi
Lennart Hoogerheide
Herman K. van Dijk
Nonlinear; non-gaussian state space; filters; density combinations; bayesian modeling; equity momentum
2016-11-17
Changes in nominal rigidities in Poland - a regime switching DSGE perspective
http://d.repec.org/n?u=RePEc:sek:iacpro:5306955&r=ore
We estimate a dynamic stochastic general equilibrium model that allows for regimes Markov switching (MS-DSGE). Existing MS-DSGE papers for the United States focus on changes in monetary policy or shocks volatility, contributing the debate on the Great Moderation and/or Volcker disinflation. However, Poland which here serves as an example of a transition country, faced a wider range of structural changes, including long disinflation, EU accession or tax changes. The model identifies high and low rigidity regimes,with the timing consistent with menu cost explanation of nominal rigidities. Estimated timing of the regimes captures the European Union accession and indirect tax changes. The Bayesian model comparison results suggest that model with switching in both analyzed rigidities is strongly favored by the data in comparison with switching only in prices or in wages. Moreover, we find significant evidence in support of independent Markov chains.
PaweÅ‚ Baranowski
Zbigniew Kuchta
nominal rigidities, Markov switching DSGE models, bayesian model comparison
Simple Forecasting Heuristics that Make us Smart: Evidence from Different Market Experiments
http://d.repec.org/n?u=RePEc:uts:ecowps:29&r=ore
We study a model in which individual agents use simple linear first order price forecasting rules, adapting them to the complex evolving market environment with a smart Genetic Algorithm optimization procedure. The novelties are: (1) a parsimonious experimental foundation of individual forecasting behaviour; (2) an explanation of individual and aggregate behavior in four different experimental settings, (3) improved one-period and 50-period ahead forecasting of lab experiments, and (4) a characterization of the mean, median and empirical distribution of forecasting heuristics. The median of the distribution of GA forecasting heuristics can be used in designing or validating simple Heuristic Switching Model.
Mikhail Anufriev
Cars Hommes
Tomasz Makarewicz
Expectation Formation; Learning to Forecast Experiment; Genetic Algorithm Model of Individual Learning
2015-07-13
On uniqueness of time-consistent Markov policies for quasi-hyperbolic consumers under uncertainty
http://d.repec.org/n?u=RePEc:sgh:kaewps:2016020&r=ore
We give a set of sufficient conditions for uniqueness of a time-consistent Markov stationary consumption policy for a quasi-hyperbolic household under uncertainty. To the best of our knowledge, this uniqueness result is the first presented in the literature for general settings, i.e. under standard assumptions on preferences, as well as some new condition on a transition probability. This paper advocates a ''generalized Bellman equation'' method to overcome some predicaments of the known methods and also extends our recent existence result. Our method also works for returns unbounded from above. We provide few natural followers of optimal policy uniqueness: convergent and accurate computational algorithm, monotone comparative statics results and generalized Euler equation.
Lukasz Balbus
Kevin Reffett
Lukasz Wozny
Time consistency, Markov equilibria, Uniqueness, Stochastic games, Generalized Bellman equation
2016-11
Markovian Nash equilibrium in financial markets with asymmetric information and related forward-backward systems
http://d.repec.org/n?u=RePEc:ehl:lserod:63259&r=ore
This paper develops a new methodology for studying continuous-time Nash equilibrium in a financial market with asymmetrically informed agents. This approach allows us to lift the restriction of risk neutrality imposed on market makers by the current literature. It turns out that, when the market makers are risk averse, the optimal strategies of the agents are solutions of a forward- backward system of partial and stochastic differential equations. In particular, the price set by the market makers solves a non-standard `quadratic' backward stochastic differential equation. The main result of the paper is the existence of a Markovian solution to this forward-backward system on an arbitrary time interval, which is obtained via a fixed-point argument on the space of absolutely continuous distribution functions. Moreover, the equilibrium obtained in this paper is able to explain several stylized facts which are not captured by the current asymmetric information models.
Umut Çetin
Albina Danilova
Kyle model with risk averse market makers; Bertrand competition; forward–backward stochastic and partial differential equations; Markov bridges
2016-09-01
Regulation and Rational Banking Bubbles in Infinite Horizon
http://d.repec.org/n?u=RePEc:luc:wpaper:16-15&r=ore
This paper develops a dynamic stochastic general equilibrium model in infinite horizon with a regulated banking sector where stochastic banking bubbles may arise endogenously. We analyze the conditions under which stochastic bubbles exist and their impact on macroeconomic key variables. We show that when banks face capital requirements based on Value-at- Risk, two different equilibria emerge and can coexist: the bubbleless and the bubbly equilibria. Alternatively, under a regulatory framework where capital requirements are based on credit risk only, as in Basel I, bubbles are explosive and, as a consequence, cannot exist. The stochastic bubbly equilibrium is characterized by positive or negative bubbles depending on the tightness of capital requirements based on Value-at-Risk. We find a maximum value of capital requirements under which bubbles are positive. Below this threshold, the stochastic bubbly equilibrium provides larger wel- fare than the bubbleless equilibrium. In particular, our results suggest that a change in banking policies might lead to a crisis without external shocks.
Claire Océane Chevallier
Sarah El Joueidi
Banking bubbles; banking regulation; DSGE; infinitely lived agents; multiple equilibria; Value-at-Risk
2016
Resource Allocation Heuristics for Unknown Sales Response Functions with Additive Disturbances
http://d.repec.org/n?u=RePEc:bay:rdwiwi:34818&r=ore
We develop an exploration-exploitation algorithm which solves the allocation of a fixed resource (e.g., a budget, a sales force size, etc.) to several units (e.g., sales districts, customer groups, etc.) with the objective to attain maximum sales. This algorithm does not require knowledge of the form of the sales response function and is also able cope with additive random disturbances. The latter as a rule are a component of sales response functions estimated by econometric methods. We compare the algorithm to three rules of thumb which in practice are often used for this allocation problem. The comparison is based on a Monte Carlo simulation for five replications of 192 experimental constellations, which are obtained from four function types, four procedures (i.e., the three rules of thumb and the algorithm), similar/varied elasticities, similar/varied saturations, and three error levels. A statistical analysis of the simulation results shows that the algorithm performs better than the three rules of thumb if the objective consists in maximizing sales across several periods. We also mention several more general marketing decision problems which could be solved by appropriate modifications of the algorithm presented.
Gahler, Daniel
Hruschka, Harald
Marketing Resource Allocation; Exploration-Exploitation Algorithm; Monte Carlo Simulation; Optimization
2016-11-17
A One-Sector Optimal Growth Model in Which Consuming Takes Time
http://d.repec.org/n?u=RePEc:mse:cesdoc:16072&r=ore
This article establishes a growth model in which consumption takes time. The agent faces a time constraint, i.e; her/his available amount of time must be optimally share between consuming time and working time. By using a dynamic programming argument, it is proved that the optimal capital sequences are monotonic and have property that converges to steady state. We also compare this model to the one agent growth model with elastic labor. We obtain that (i) When the quantity of time to consume one unit of consumption increases, the agent devotes less time for labour. (ii) When the quantity of time to consume one unit of consumption is smaller that the threshod, it is better for the economy to spend time to consume than to enjoy leisure. We have more time for labour. This implies more output and more consumption. We reverse the situation when the quantity of time to consume one unit of consumption is larger than the threshold. We give an example to illustrate this result. Finally, if both models have the same technology which is of constant returns to scale, then they have the same ratios capital stock per head and consumption per head
Cuong Le Van
Thai Ha-Huy
Thi-Do-Hanh Nguyen
time consuming model, allocation of time; elastic labour; leisure; value function
2016-10
A New Nonlinearity Test to Circumvent the Limitation of Volterra Expansion with Applications
http://d.repec.org/n?u=RePEc:pra:mprapa:75216&r=ore
In this paper, we propose a quick, efficient, and easy method to examine whether a time series Yt possesses any nonlinear feature. The advantage of our proposed nonlinearity test is that it is not required to know the exact nonlinear features and the detailed nonlinear forms of Yt. We find that our proposed test can be used to detect any nonlinearity for the variable being examined and detect GARCH models in the innovations. It can also be used to test whether the hypothesized model, including linear and nonlinear, to the variable being examined is appropriate as long as the residuals of the model being used can be estimated. Our simulation study shows that our proposed test is stable and powerful. We apply our proposed statistic to test whether there is any nonlinear feature in the sunspot data and whether the S&P 500 index follows a random walk model. The conclusion drawn from our proposed test is consistent those from other tests.
Hui, Yongchang
Wong, Wing-Keung
Bai, Zhidong
Zhu, Zhenzhen
Nonlinearity, U-statistics, Volterra expansion, sunspots, efficient market
2016-11-22
Portfolio analysis in jump-diffusion model with power-law tails
http://d.repec.org/n?u=RePEc:sek:iacpro:5306873&r=ore
The classic portfolio analysis given by Markowitz theory and Capital Asset Pricing Model is based on the assumption that the assetsâ€™ returns are normally distributed. In this situation one can use only two criteria: expected return and variance of return as the measures of possible gains and risk, respectively. However there is a growing evidence that the assetsâ€™ returns and in particular returns of shares in the stock markets fail to obey Gaussian distribution. Therefore different measures of risk should be considered.In the paper we analyze the portfolio problem in the situation when stock prices follows jump-diffusion model with the tails of jumps obeying power-law. We consider a portfolio problem with two risk criteria: risk in the situation of normal market circumstances and the risk of jumps. We propose a method for numerical computing the former risk using Fast Fourier Transform (FFT). Finally we present the examples of portfolio analysis with the new method for the shares from Warsaw Stock Market Exchange.
PaweÅ‚ Kliber
portfolio analysis, jump-diffusion models, power-law, risk of extremes, Fast Fourier Transform