I would like to solve constrained maximization problem by the Lagrangian.

$ max_{q(\omega)}\ U=\left(\int_{0}^{n} q(\omega)^{\rho} d \omega\right)^{\frac{1}{\rho}} \quad 0<\rho<1 $

$s.t.\ \int_{0}^{n} p(\omega) q(\omega) d \omega = w$

Setting the Lagrangian function, $\mathcal{L}=U^{\rho}-\lambda\left(\int_{0}^{n} p(\omega) q(\omega) d \omega-\mathrm{w}\right)$

Take the first derivatives, $\frac{\partial \mathcal{L}}{\partial q(\omega)}=\rho q(\omega)^{\rho-1}-\lambda p(\omega)=0$

Rearranging terms yields, $q(\omega)=\left(\frac{\lambda p(\omega)}{\rho}\right)^{\frac{1}{\rho-1}}$

I try the following code

U[i_] := Integrate[q[i]^rho, {i, 0, n}]^(1/rho)

L[n_, q_, p_] := 
 U[n]^rho - lambda*(Integrate[q[i]*p[i], {i, 0, n}

Solve[D[L[n, q, p], p]==0,q]

that is not working.

  • 1
    $\begingroup$ Exponent is p not $\rho$. What is utility? $\endgroup$ Jun 18, 2020 at 4:17
  • 1
    $\begingroup$ A typo. It is U. @OkkesDulgerci $\endgroup$
    – XJ.C
    Jun 18, 2020 at 4:31
  • $\begingroup$ Derivative should be respect to q not p $\endgroup$ Jun 18, 2020 at 4:43

1 Answer 1

U := Integrate[q^ρ, {w, 0, n}]^(1/ρ)
L[w_, q_, p_] :=  PowerExpand[U^ρ] - λ (Integrate[p q, {w, 0, n}] - w);

Solve[D[L[w, q, p], q] == 0, q] // Quiet

$\left\{q\to \left(\frac{\lambda p}{\rho }\right)^{\frac{1}{\rho -1}}\right\}$


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

Not the answer you're looking for? Browse other questions tagged or ask your own question.