scispace - formally typeset
Open AccessJournal ArticleDOI

Differentiability of the value function without interiority assumptions

Reads0
Chats0
TLDR
Under minimal assumptions it is proved that the value function is continuously differentiable in concave dynamic programs.
About
This article is published in Journal of Economic Theory.The article was published on 2009-09-01 and is currently open access. It has received 57 citations till now. The article focuses on the topics: Differentiable function & Semi-differentiability.

read more

Citations
More filters
Journal ArticleDOI

On the smoothness of value functions and the existence of optimal strategies in diffusion models

TL;DR: For time-homogeneous optimal control problems with a one-dimensional diffusion, it is proved that the corresponding value function must be twice continuously differentiable under Lipschitz, growth, and non-vanishing-volatility conditions.
Journal ArticleDOI

Approximate dynamic programming for stochastic N-stage optimization with application to optimal consumption under uncertainty

TL;DR: Both theoretical and numerical results give insights into the possibility of coping with the curse of dimensionality in stochastic optimization problems whose decision strategies depend on large numbers of variables.

Numerical simulation of nonoptimal dynamic equilibrium models

TL;DR: In this article, a recursive method for the computation of dynamic competitive equilibria in models with heterogeneous agents and market frictions is presented, based on a convergent operator over an expanded set of state variables.
Journal ArticleDOI

Turning from Crime: A Dynamic Perspective *

TL;DR: In this article, a variant of the human capital model is used to examine criminal choice using a variation of human capital, which disaggregates an individual's capital stock into the standard human capital component as well as a utility generating component, called social capital.
Journal ArticleDOI

Inequality Constraints and Euler Equation-based Solution Methods

TL;DR: It is shown that a common iterative procedure on the Euler equation delivers a sequence of approximate policy functions that converges to the true solution under a wide range of circumstances.
References
More filters
Book

Optimization and nonsmooth analysis

TL;DR: The Calculus of Variations as discussed by the authors is a generalization of the calculus of variations, which is used in many aspects of analysis, such as generalized gradient descent and optimal control.
Journal ArticleDOI

Asset prices in an exchange economy

Robert E. Lucas
- 01 Nov 1978 - 
TL;DR: In this article, the authors examine the stochastic behavior of equilibrium asset prices in a one-good, pure exchange economy with identical consumers, and derive a functional equation for price as a function of the physical state of the economy.
Book

Recursive methods in economic dynamics

TL;DR: In this article, a deterministic model of optimal growth is proposed, and a stochastic model is proposed for optimal growth with linear utility and linear systems and linear approximations.
Book

Recursive Macroeconomic Theory

TL;DR: In this paper, an introduction to recursive methods for dynamic macroeconomics is presented, including standard applications such as asset pricing, and advanced material, including analyses of reputational mechanisms and contract design.
Journal ArticleDOI

Envelope Theorems for Arbitrary Choice Sets

TL;DR: The standard envelope theorems apply to choice sets with convex and topological structure, providing sufficient conditions for the value function to be differentiable in a parameter and characterizing its derivative as mentioned in this paper.
Frequently Asked Questions (17)
Q1. What are the contributions in "Differentiability of the value function without interiority assumptions" ?

This paper studies first-order differentiability properties of the value function in concave dynamic programs. Under minimal assumptions the authors prove that the value function is continuously differentiable. The authors then discuss this result in the context of some economic models, and focus on some examples in which their assumptions are not met and the value function is not differentiable. 

For each fixed z Z mapping Q(z, ·) : Z R is a probability measure, and for each fixed B Z mapping Q(·,B) :Z R is a measurable function. 

By the welfare theorems, the uniqueness of the multipliers entails that an optimal allocation is just supported by a unique price system. 

To establish that function v is of class C1, the authors require differentiability of the return function U , and some regularity conditions for boundary solutions. 

The authors consider two conditions transported from the static theory: A boundary restriction on the policy function over the domain of definition and a constraint qualification (cf. [20]). 

In their case, the following conditions ensure this contraction property: (i) Function U(x, y, z) and its partial derivatives DjU(x, y, z), j = 1,2, are uniformly bounded on × Z, and (ii) sup(x,y,z) ∫ Z G(x, y, z )μ(z, dz ) < 1. 

Uaut for all s, (P3) Us [Uaut,Umax] for all s,where the value function V (U0) assigns the maximum utility to an agent (say agent 2) over all possible utility levels U0 of the other agent. 

The recursive formulation of the pure currency model isv(m, z) = max c, m 0{ u(c, z) + ∫ Z v(m , z )μ(dz ) }s.t. m (1 + ) + c m y 0, c m 0. 

The non-negativity conditions of Proposition 3.2 allow for a simple extension of the transversality condition to unbounded domains, e.g., see Benveniste and Scheinkman [3]. 

The authors show that in the above growth model studied in Section 3.4.2, the derivative of the value function may become unbounded as the stock of capital approaches zero – even if the derivatives of the utility and production functions are bounded. 

The sequence of choice variables {xt }t 0 belongs to a set X Rn, and the realizations of the exogenous stochastic process {zt }t 0 lie in a space Z. Let X be the Borel -algebra of X and Z the -algebra of Z. 

A contingency plan {xt }t 0 is feasible if xt+1 :Zt X is a measurable function and xt+1(zt ) (xt (zt 1), zt ), for zt Zt and t = 0,1, . . . .The stochastic optimization problem can be defined as follows: 

The decomposition given in Theorem 3.1 of the derivative of the value function into the fundamental value and a bubble term is common in asset pricing, where B0 can be identified with the bubble term of some existing assets. 

The interpretation of (3) is as follows: For z Z fixed, qz1(Mv)(y, z) if and only if there is a measurable mapping z qz(z ) with qz(z ) 1v(y, z ) almost everywhere (a.e.) in the measure Q(z, ·) such that qz = ∫ Z qz(z )Q(z, dz ). 

Then for each (x, z) there exists some optimal solution y H(x, z) with y int(X) such that the rank of the matrix of partial derivatives {D2gi(x, y, z): i The author(x, y, z)} is equal to s(x, y, z). 

D1: For every x int(X) and z Z function U(·,·, z) is of class C1 on some open neighborhood N(x,y) of every point (x, y) with y H(x, z). 

End-of-period asset holdings can be defined in a rather arbitrary way, as the agent can be endowed with new securities at the beginning of each period so as to replicate the optimal path {xt+1(zt )}t 0.