7.3 Optimization with inequality constraints: the sufficiency of the Kuhn-Tucker conditions
Concave objective function
- Proposition 7.3.1 (Sufficiency of Kuhn-Tucker conditions for concave objective function)
-
Let f and gj for j = 1, ..., m be continuously differentiable functions of many variables defined on an open convex set S and let cj for j =
1, ..., m be constants. Suppose that
- f is concave
- and gj is quasiconvex for j = 1, ..., m.
maxx∈Sf(x) subject to gj(x) ≤ cj for j = 1, ..., m.
- Proof
-
The fact that (x*, λ) satisfies the Kuhn-Tucker conditions means that
f'i(x*) − ∑m
j=1λjg'ji(x*) = 0 for i = 1, ..., nλj ≥ 0, gj(x*) ≤ cj and λj[gj(x*) − cj] = 0 for j = 1, ..., m Let x be such that gj(x) ≤ cj for all j. We need to show that f(x) ≤ f(x*).
Let j be an index such that gj(x*) = cj. (That is, the jth constraint is binding at x*.) Then gj(x) ≤ gj(x*), so that by a characterization of quasiconvex functions,
∑n 0. Thus given λj ≥ 0 we have
i=1g'ji(x*)·(xi − x*i) ≤λj∑n 0.
i=1g'ji(x*)·(xi − x*i) ≤Now let j be an index such that gj(x*) < cj. Then by the complementary slackness condition for constraint j in the Kuhn-Tucker conditions, λj = 0.
Thus for every constraint j we have
λj∑n
i=1g'ji(x*)·(xi − x*i) ≤ 0∑m
j=1λj∑n
i=1g'ji(x*)·(xi − x*i) ≤ 0,∑n
i=1∑m
j=1λjg'ji(x*)·(xi − x*i) ≤ 0.∑n
i=1f'i(x*)·(xi − x*i) − ∑n
i=1∑m
j=1λjg'ji(x*)·(xi − x*i) = 0.∑n
i=1f'i(x*)·(xi − x*i) ≤ 0.
- Corollary 7.3.1 (Necessity and sufficiency of Kuhn-Tucker conditions for concave objective function)
-
Let f and gj for j = 1, ..., m be continuously differentiable functions of many variables defined on an open convex set S and let cj for j =
1, ..., m be constants. Suppose that f is concave and
- either each gj is linear
- or each gj is convex and there exists x ∈ S such that gj(x) < cj for j = 1, ..., m.
maxx∈Sf(x) subject to gj(x) ≤ cj for j = 1, ..., mif and only if there exists an m-vector of numbers λ = (λ1, ..., λm) such that (x*, λ) satisfies the Kuhn-Tucker conditions.
Quasiconcave objective function
The condition that the objective function be concave is a bit too strong to be useful in some economic applications. For example, the assumption we would like to impose on a consumer's utility function is that it be quasiconcave. The next result is useful in this case.- Proposition 7.3.2 (Sufficiency of Kuhn-Tucker conditions for quasiconcave objective function)
-
Let f and gj for j = 1, ..., m be continuously differentiable functions of n variables defined on an open convex set S and let cj for j =
1, ..., m be constants. Suppose that
- f is quasiconcave
- and gj is quasiconvex for j = 1, ..., m.
maxx∈Sf(x) subject to gj(x) ≤ cj for j = 1, ..., m.
- Proof
-
The proof of the previous result, Proposition 7.3.1, uses the concavity of f only in the last sentence. In particular, under the assumption that f is quasiconcave we can conclude that
∑n
i=1f'i(x*)·(xi − x*i) ≤ 0,Denote the vector of partial derivatives of f by ∇f. Given the continuity of f, the fact that f(x) > f(x*) means that there exists t > 0 (small enough) such that f(x − t∇f(x*)) > f(x*), and the conclusion from the proof of the Proposition 7.3.1 may be written
∇f(x*)·(x − x*) ≤ 0. ∇f(x*)·(x − t∇f(x*) − x*) = −t∇f(x*)·∇f(x*) + ∇f(x*)(x − x*) ≤ −t∑n
i=1(f'i(x*))2 < 0,
- Corollary 7.3.2 (Necessity and sufficiency of Kuhn-Tucker conditions for quasiconcave objective function)
-
Let f and gj for j = 1, ..., m be continuously differentiable functions of n variables defined on an open convex set S and let cj for j =
1, ..., m be constants. Suppose that f is quasiconcave and each gj is linear.
If x* ∈ S solves the problem
maxx∈Sf(x) subject to gj(x) ≤ cj for j = 1, ..., mthen there exists a unique m-vector λ such that (x*, λ) satisfies the Kuhn-Tucker conditions, and if there exists an m-vector λ and a point x* ∈ S such that (x*, λ) satisfies the Kuhn-Tucker conditions and it is not the case that f'i(x*) = 0 for i = 1, ..., n then x* solves the problem.
Examples
The next two simple examples illustrate how to use these results and the ones in Section 7.2 to solve optimization problems.- Example 7.3.1
-
Consider the problem
maxx∈S[−(x − 2)2] subject to x ≥ 1,where S is the set of all numbers. This problem is illustrated in the following figure.
Written in the standard format, this problem is
maxx∈S[−(x − 2)2] subject to 1 − x ≤ 0.The set S is open and convex, and the objective and constraint functions are continuously differentiable on this set. The objective function is concave and the constraint function is linear, and hence both concave and quasiconvex. Thus by Proposition 7.2.1 the Kuhn-Tucker conditions are necessary (if x* solves the problem then there is a number λ such that (x*, λ) satisfies the Kuhn-Tucker conditions) and by Proposition 7.3.1 they are sufficient (if (x*, λ) satisfies the Kuhn-Tucker conditions then x* solves the problem).The Lagrangean is the function L defined by
L(x) = −(x − 2)2 − λ(1 − x).The Kuhn-Tucker conditions are
−2(x − 2) + λ = 0 1 − x ≤ 0, λ ≥ 0, and λ(1 − x) = 0. - x = 1: 2 + λ = 0, or λ = −2, which violates λ ≥ 0.
- λ = 0: −2(x − 2) = 0; the only solution is x = 2.
- Example 7.3.2
-
Consider the problem
maxx∈S[−(x − 2)2] subject to x ≥ 3,where S is the set of all numbers. This problem is illustrated in the following figure.
Written in the standard format, this problem is
maxx∈S[−(x − 2)2] subject to 3 − x ≤ 0.As in the previous example, 7.3.1, the objective function is concave and the constraint function is linear, so that x* is a solution of the problem if and only if there is a number λ such that (x*, λ) is a solution of the Kuhn-Tucker conditions.The Lagrangean is
L(x) = −(x − 2)2 − λ(3 − x).The Kuhn-Tucker conditions are
−2(x−2) + λ = 0 3 − x ≤ 0, λ ≥ 0, and λ(3 − x) = 0. - x = 3: −2 + λ = 0, or λ = 2.
- λ = 0: −2(x − 2) = 0; since x ≥ 3 this has no solution compatible with the other conditions.
- Example 7.3.3
-
Consider the problem
max(x1, x2)∈S [−(x1 − 4)2 − (x2 − 4)2] subject to x1 + x2 ≤ 4 and x1 + 3x2 ≤ 9,where S is the set of all pairs of numbers. The set S is open and convex, and the objective and constraint functions are continuously differentiable on this set. The objective function is concave and each constraint function is linear, and hence both concave and quasiconvex. Thus by Proposition 7.2.1 the Kuhn-Tucker conditions are necessary (if x* solves the problem then there is a number λ such that (x*, λ) satisfies the Kuhn-Tucker conditions) and by Proposition 7.3.1 they are sufficient (if (x*, λ) satisfies the Kuhn-Tucker conditions then x* solves the problem).
In Example 7.1.1 we previously found that the Kuhn-Tucker conditions for this problem are
−2(x1 − 4) − λ1 − λ2 = 0 −2(x2 − 4) − λ1 − 3λ2 = 0 x1 + x2 ≤ 4, λ1 ≥ 0, and λ1(x1 + x2 − 4) = 0 x1 + 3x2 ≤ 9, λ2 ≥ 0, and λ2(x1 + 3x2 − 9) = 0. - x1 + x2 = 4 and x1 + 3x2 = 9: In this case we have x1 = 3/2 and x2 = 5/2. Then the first two equations are
5 − λ1 − λ2 = 0 3 − λ1 − 3λ2 = 0 - x1 + x2 = 4 and x1 + 3x2 < 9, so that λ2 = 0: Then first two equations imply x1 = x2 = 2 and λ1 = 4. All the conditions are satisfied, so (x1, x2, λ1, λ2) = (2, 2, 4, 0) is a solution.
- x1 + x2 < 4 and x1 + 3x2 = 9, so that λ1 = 0: Then the first two equations imply x1 = 33/10 and x2 = 19/10, violating x1 + x2 < 4.
- x1 + x2 < 4 and x1 + 3x2 < 9, so that λ1 = λ2 = 0: Then first two equations imply x1 = x2 = 4, violating x1 + x2 < 4.
- x1 + x2 = 4 and x1 + 3x2 = 9: In this case we have x1 = 3/2 and x2 = 5/2. Then the first two equations are
- Example 7.3.4
-
Consider the problem
max(x,y)∈S xy subject to x + y ≤ 6, x ≥ 0, and y ≥ 0,where S is the set of all pairs of numbers. This set is open and convex, and the objective and constraint functions are differentiable on it. Each constraint function is linear, and hence concave. Thus by Proposition 7.2.1 the Kuhn-Tucker conditions are necessary (if x* solves the problem then there is a vector λ such that (x*, λ) satisfies the Kuhn-Tucker conditions). Also, the objective function is continuous and the constraint set is compact, so by the extreme value theorem the problem has a solution. Thus x* is a solution of the problem if and only if there is a vector λ* such that (x*, λ*) satisfies the Kuhn-Tucker conditions and f(x*) ≥ f(x) for all values of x for which there is a vector λ such that (x, λ) satisfies the Kuhn-Tucker conditions.
(Note that we cannot use Proposition 7.3.1 because the objective function is not concave, and we cannot use Proposition 7.3.2 because although the objective function is quasiconcave on the set of vectors x with x ≥ 0, it is not quasiconcave on any open set that contains this set.)
The Lagrangean is given by
L(x, y) = xy − λ1(x + y − 6) + λ2x + λ3y.The Kuhn-Tucker conditions arey − λ1 + λ2 = 0 x − λ1 + λ3 = 0 λ1 ≥ 0, x + y ≤ 6, λ1(x + y − 6) = 0 λ2 ≥ 0, x ≥ 0, λ2x = 0 λ3 ≥ 0, y ≥ 0, λ3y = 0. - If x > 0 and y > 0 then λ2 = λ3 = 0, so that λ1 = x = y from the first two conditions. Hence x = y = λ1 = 3 from the third condition. These values satisfy all the conditions.
- If x = 0 and y > 0 then λ3 = 0 from the last condition and hence λ1 = x = 0 from the second condition. But now from the first condition λ2 = −y < 0, contradicting λ2 ≥ 0.
- If x > 0 and y = 0 then λ2 = 0, and a symmetric argument yields a contradiction.
- If x = y = 0 then λ1 = 0 from the third set of conditions, so that λ2 = λ3 from the first and second conditions. These values satisfy all the conditions.
- Example 7.3.5
-
Consider the problem
max(x,y)∈S x1/2 + y subject to px + y ≤ I, x ≥ 0, and y ≥ 0where S is the set of pairs (x, y) of numbers with x ≥ 0, and p > 0 and I > 0 are parameters. The set S is not open, and the objective function is defined only for x ≥ 0, so it is not defined on any open set that contains the constraint set. Thus the argument in the previous example cannot be used.
Instead, first consider the problem
max(x,y)∈S' x1/2 + y subject to px + y ≤ I and y ≥ 0where S' = {(x, y): x > 0}. The set S' is open and convex, the objective and constraint functions are continuously differentiable on this set, the objective function is quasiconcave, and each constraint function is linear, so by Corollary 7.3.2 if x* ∈ S' solves the problem then there is a vector λ such that (x*, λ) satisfies the Kuhn-Tucker conditions and if there is a vector λ and a point x* ∈ S' such that (x*, λ) satisfies the Kuhn-Tucker conditions and not all the partial derivatives of the objective function are 0 at x* then x* solves the problem.The Lagrangean of the modified problem is given by
L(x, y) = x1/2 + y − λ1(px + y − I) + λ2y.The Kuhn-Tucker conditions are(1/2)x−1/2 − λ1p = 0 1 − λ1 + λ2 = 0 λ1 ≥ 0, px + y ≤ I, and λ1(px + y − I) = 0 λ2 ≥ 0, −y ≤ 0, and λ2y = 0. - If λ2 = 0 then λ1 = 1 from the second condition, and hence x = 1/(4p2) and y = I − 1/(4p). We have 1/(4p2) > 0, so (x, y, λ1, λ2) = (1/(4p2), I − 1/(4p), 1, 0) is a solution of the Kuhn-Tucker conditions if I − 1/(4p) ≥ 0, or p ≥ 1/(4I).
- If y = 0 then x = I/p, λ1 = 1/(2(pI)1/2), and λ2 = λ1 − 1, so that λ2 ≥ 0 if and only if λ1 ≥ 1, or p ≤ 1/(4I).
(I/p, 0) if p ≤ 1/(4I) (1/(4p2),I − 1/(4p)) if p > 1/(4I). Now return to the original problem. The maximal value of the objective function when x = 0 is I. At the solution of the modified problem, the value of the objective function is (I/p)1/2 if p ≤ 1/(4I), and 1/(2p) + I − 1/(4p) = I + 1/(4p) if p > I/(4p). If p ≤ 1/(4I) then (I/p)1/2 ≥ 2I, so in both cases the value of the objective function exceeds I. We conclude that the solution of the original problem is the solution of the modified problem.