Economics - Research Publications

Permanent URI for this collection

http://hdl.handle.net/11343/417

Filters

2004 - 2006 7
Reset filters

Settings
Statistics
Citations
Start date: 2003 × End date: 2010 × Type: Preprint × Author: STACHURSKI, JOHN ×

Search Results

Now showing 1 - 7 of 7
  • Item
    Thumbnail Image
    Stochastic growth with nonconvexities: the optimal case
    Nishimura, Kazuo ; Rudnicki, Ryszard ; STACHURSKI, JOHN ( 2004-02)
    This paper studies optimal investment and dynamicbehaviour of stochastically growing economies. We assume neitherconvex technology nor bounded support of the productivity shocks.A number of basic results concerning the investment policy and theRamsey–Euler equation are established. We also prove a fundamentaldichotomy pertaining to optimal growth models perturbedby standard econometric shocks: Either an economy is globallystable or it is globally collapsing to the origin.
  • Item
    Thumbnail Image
    Asymptotic statistical properties of the neoclassical optimal growth model
    STACHURSKI, JOHN ( 2004-02)
    The standard one-sector stochastic optimal growthmodel is shown to be not just ergodic but geometrically ergodic.In addition, it is proved that the time series generated by the optimalpath satisfy the Law of Large Numbers and the Central LimitTheorem.
  • Item
    Thumbnail Image
    Equivalent conditions for irreducibility of discrete time Markov chains
    Van, Cuong Le ; STACHURSKI, JOHN ( 2004-02)
    We consider discrete time Markov chains on generalstate space. It is shown that a certain property referred to hereas nondecomposability is equivalent to irreducibility, and that aMarkov chain with invariant distribution is irreducible if and only ifthe invariant distribution is unique and assigns positive probabilityto all absorbing sets.
  • Item
    Thumbnail Image
    Stochastic optimal growth when the discount rate vanishes
    Nishimura, Kazuo ; STACHURSKI, JOHN ( 2004-07)
    It has been shown that long-run optimality of thelimit of discounted optima when the discount rate vanishes is impliedby a condition on the value function of the optimal program.We suggest a new method to verify this condition in the contextof one-sector optimal growth. The idea should be more widelyapplicable.
  • Item
    Thumbnail Image
    Computable Bounds for ExtremeEvent Probabilities in StochasticEconomic Models
    STACHURSKI, JOHN ( 2005-02)
    The paper introduces a multiplicative drift conditionfor evaluating stochastic economic models. The drift condition isshown to permit computation of quantitative bounds for extremeevent probabilities in terms of the model primitives. By way ofillustration, the technique is applied to a simple threshold autoregressionmodel of exchange rates.
  • Item
    Thumbnail Image
    Computing the Distributions of EconomicModels Via Simulation
    STACHURSKI, JOHN ( 2005-10)
    This paper studies the convergence properties of aMonte Carlo algorithm for computing distributions of state variableswhen the underlying model is a Markov chain with absolutelycontinuous transition probabilities. We show that the L1 error ofthe estimator always converges to zero with probability one. Inaddition, rates of convergence are established for L1 and integralmean squared errors. The algorithm is shown to have many applicationsin economics.
  • Item
    Thumbnail Image
    Continuous state dynamic programming via nonexpansive approximation
    STACHURSKI, JOHN ( 2006-02)
    This paper studies fitted value iteration for continuousstate dynamic programming using nonexpansive function approximators.A number of nonexpansive approximation schemesare discussed. The main contribution is to provide error boundsfor approximate optimal policies generated by the value iterationalgorithm.