Mathematical methods for economic theory

Martin J. Osborne

6.2 Optimization with equality constraints: n variables, m constraints

The Lagrangean method can easily be generalized to a problem of the form
maxx f(x) subject to gj(x) = cj for j = 1, ..., m
with n variables and m constraints (where x = (x1, ..., xn)).

The Lagrangean for this problem is

L(x) = f(x) − ∑m
j=1
λj(gj(x) − cj).
That is, there is one Lagrange multiplier for each constraint.

As in the case of a problem with two variables and one constraint, the first-order condition is that x* be a stationary point of the Lagrangean. The “nondegeneracy” condition in the two variable case—namely that at least one of g'1(x1x2) and g'2(x1x2) is nonzero—is less straightforward to generalize. The appropriate generalization involves the Jacobian matrix of the constraint functions (g1, ..., gm), named in honor of Carl Gustav Jacob Jacobi (1804–1851) and defined as follows.

Definition
For j = 1, ..., m let gj be a differentiable function of n variables. The Jacobian matrix of (g1, ..., gm) at the point x is
left parenthesis (∂g1/∂x1)(x) ... (∂g1/∂xn)(x) right parenthesis
... ... ...
(∂gm/∂x1)(x) ... (∂gm/∂xn)(x)
.
Proposition 6.2.1 (Necessary conditions for an extremum)  
Let f and gj for j = 1, ..., m be functions of n variables defined on a set S, with m ≤ n, that are continuously differentiable on the interior of S, let cj for j = 1, ..., m be numbers, and suppose that x* is an interior point of S that solves the problem
maxx f(x) subject to gj(x) = cj for j = 1, ..., m
or the problem
minx f(x) subject to gj(x) = cj for j = 1, ..., m
or is a local maximizer or minimizer of f(x) subject to gj(x) = cj for j = 1, ..., m. Suppose also that the rank of the Jacobian matrix of (g1, ..., gm) at the point x* is m.

Then there exist unique numbers λ1, ..., λm such that x* is a stationary point of the Lagrangean function L defined by

L(x) = f(x) − ∑m
j=1
λj(gj(x) − cj).
That is, x* satisfies the first-order conditions
L'i(x*)  =  fi'(x*) − m
j=1
λj(∂gj/∂xi)(x*) =
0 for i = 1, ..., n.
In addition, gj(x*) = cj for j = 1, ..., m.
Source  
For proofs, see Sydsæter (1981), Theorem 5.20 (p. 275) and Simon and Blume (1994), pp. 478–480. (Only Sydsæter argues explicitly that the Lagrange multipliers are unique.)
As in the case of a problem with two variables and one constraint, an implication of this result is the following procedure.
Procedure for solving a maximization problem with many equality constraints
Let f and gj for j = 1, ..., m be functions of n variables defined on a set S, with m ≤ n, that are continuously differentiable on the interior of S, and let cj for j = 1, ..., m be numbers. If the problem
maxx f(x) subject to gj(x) = cj for j = 1, ..., m
has a solution, it may be found as follows.
  • Find all the values of (x, λ) (where λ = (λ1, ..., λm)) for which (a) x is an interior point of S and (b) (x, λ) satisfies the first-order conditions given in the previous result and the constraints.
  • Find all the points x for which the rank of the Jacobian matrix of (g1, ..., gm) is less than m and gj(x) = cj for j = 1, ..., m.
  • If the set S has any boundary points, find all the points that solve the problem maxxf(x) subject to the conditions that gj(x) = cj for j = 1, ..., m and x is a boundary point of S.
  • The points x you have found for which f(x) is largest are the solutions of the problem.
Also as in the case of a problem with two variables and one constraint, the first-order conditions and the constraint are sufficient for a maximum if the Lagrangean is concave, and are sufficient for a minimum if the Lagrangean is convex, as stated precisely in the following result. The proof is the same as the proof for the result for two variables and a single constraint.
Proposition 6.2.2 (Conditions under which necessary conditions for an extremum are sufficient)
Let f and gj for j = 1, ..., m be functions of n variables defined on a convex set S that are continuously differentiable on the interior of S. Suppose that there are numbers λ*1, ..., λ*m and an interior point x* of S such that x* is a stationary point of the Lagrangean
L(x) = f(x) − ∑m
j=1
λ*j (gj(x) − cj).
Suppose further that gj(x*) = cj for j = 1, ..., m. Then
  • if L is concave—in particular if f is concave and λ*j gj is convex for j = 1, ..., m—then x* solves the constrained maximization problem
  • if L is convex—in particular if f is convex and λ*j gj is concave for j = 1, ..., m—then x* solves the constrained minimization problem
Example 6.2.1
Consider the problem
minx,y,z x2 + y2 + z2 subject to x + 2y + z = 1 and 2x − y − 3z = 4
with the domain of the objective function the set of all triples (xyz) of real numbers, which has no boundaries.

The Lagrangean is

L(xyz)  =  x2 + y2 + z2 − λ1(x + 2y + z − 1) − λ2(2x − y − 3z − 4).
The Jacobian matrix of the constraints is
left parenthesis 1 2 1 right parenthesis
2 −1 −3
which has rank 2, so any solution of the problem is a stationary point. Thus the set of solutions of the problem coincides with the set of stationary points.

The Lagrangean is convex for any values of λ1 and λ2, so that any stationary point is a solution of the problem.

The first-order conditions are

2x − λ1 − 2 λ2  =  0
2y − 2 λ1 + λ2  =  0
2z − λ1 + 3 λ2  =  0
and the constraints are
x + 2 y + z  =  1
2 x − y − 3 z  =  4
Solve first two first-order conditions for λ1 and λ2 to give
λ1  =  (2/5)x + (4/5)y
λ2  =  (4/5)x − (2/5)y.
Now substitute into the last first-order condition and then use the constraints to get
x = 16/15, y = 1/3, z = −11/15,
with λ1 = 52/75 and λ2 = 54/75.

We conclude that (xyz) = (16/15, 1/3, −11/15) is the unique solution of the problem.

Interpretation of Lagrange multipliers

In the case of a problem with two variables and one constraint we saw that the Lagrange multiplier is equal to the rate of increase of the maximal value of the objective function as the constraint is relaxed. This result has a natural generalization to problems with n variables and m constraints.

Consider the problem

maxx f(x) subject to gj(x) = cj for j = 1, ..., m,
where x = (x1, ..., xn). For each value of c = (c1, ..., cm), let x*(c) be a solution of this problem that is interior to the domain of the objective function, and assume that the Jacobian matrix of (g1, ..., gm) at x* has rank m. Then by the first result on this page, there are numbers λ1(c), ..., λm(c) such that x*(c) satisfies the first-order conditions given in that result.

Let

f*(c) = f(x*(c)),
the maximized value of the objective function. Then if the functions x* and λ1, ..., λm are differentiable, we can show that
f*j'(c) = λj(c) for j = 1, ..., m.

That is:

if the solution of the problem is an interior point of the domain of the objective function that satisfies the first-order conditions and is a differentiable function of c, then the value of the Lagrange multiplier for the jth constraint at the solution is equal to the rate of change in the maximal value of the objective function as cj increases.
If the jth constraint arises because of a limit on the amount of some resource, we sometimes call λj(c) the shadow price of the jth resource.