mathematics
play

MATHEMATICS 1 CONTENTS More than two variables More than one - PowerPoint PPT Presentation

Multiple constrained optimization BUSINESS MATHEMATICS 1 CONTENTS More than two variables More than one constraint Lagrange method The Lagrange multiplier Old exam question Further study 2 MORE THAN TWO VARIABLES Recall the constrained


  1. Multiple constrained optimization BUSINESS MATHEMATICS 1

  2. CONTENTS More than two variables More than one constraint Lagrange method The Lagrange multiplier Old exam question Further study 2

  3. MORE THAN TWO VARIABLES Recall the constrained optimization problem: α‰Š max 𝑔 𝑦, 𝑧 s.t. 𝑕 𝑦, 𝑧 = 𝑑 β–ͺ function to be maximized depends on 𝑦 and 𝑧 But sometimes there are more variables β–ͺ 𝑔 𝑦, 𝑧, 𝑨 or 𝑔 𝑦 1 , 𝑦 2 , 𝑦 3 3

  4. MORE THAN TWO VARIABLES Problem formulation: α‰Š max 𝑔 𝑦, 𝑧, 𝑨 s.t. 𝑕 𝑦, 𝑧, 𝑨 = 𝑑 Constrained optimization problem with more than two variables In vector notation: α‰Š max 𝑔 𝐲 s.t. 𝑕 𝐲 = 𝑑 4

  5. EXERCISE 1 Write the following in standard notation: constraint 𝑦 = 2𝑧𝑨 , objective function 𝑦𝑧 2 + 𝑨 , purpose is maximization. 5

  6. MORE THAN ONE CONSTRAINT Again recall the constrained optimization problem: α‰Š max 𝑔 𝑦, 𝑧, 𝑨 s.t. 𝑕 𝑦, 𝑧, 𝑨 = 𝑑 β–ͺ constraint is 𝑕 𝑦, 𝑧, 𝑨 = 𝑑 But sometimes there are more constraints β–ͺ β„Ž 𝑦, 𝑧, 𝑨 = 𝑒 or 𝑕 2 𝑦, 𝑧, 𝑨 = 𝑑 2 7

  7. MORE THAN ONE CONSTRAINT Problem formulation: max 𝑔 𝑦, 𝑧, 𝑨 s.t. 𝑕 1 𝑦, 𝑧, 𝑨 = 𝑑 1 ࡞ and 𝑕 2 𝑦, 𝑧, 𝑨 = 𝑑 2 Constrained optimization problem with more than one constraint In vector notation: α‰Š max 𝑔 𝐲 s.t. 𝐑 𝐲 = 𝐝 8

  8. EXERCISE 2 Write the following in vector notation: constraints 𝑦 = 2𝑧𝑨 and 𝑦 2 βˆ’ 5 = 𝑨 , objective function 𝑦𝑧 2 + 𝑨 , purpose is maximization. 9

  9. LAGRANGE METHOD How to solve the problem with three variables? α‰Š max 𝑔 𝑦, 𝑧, 𝑨 s.t. 𝑕 𝑦, 𝑧, 𝑨 = 𝑑 Generalized trick: β–ͺ introduce Lagrange multiplier πœ‡ β–ͺ define Lagrangian β„’ 𝑦, 𝑧, 𝑨, πœ‡ β„’ 𝑦, 𝑧, 𝑨, πœ‡ = 𝑔 𝑦, 𝑧, 𝑨 βˆ’ πœ‡ 𝑕 𝑦, 𝑧, 𝑨 βˆ’ 𝑑 Find the stationary points β–ͺ by solving four equations: πœ–β„’ πœ–β„’ πœ–β„’ πœ–β„’ πœ–π‘¦ = 0 , πœ–π‘§ = 0 , πœ–π‘¨ = 0 , and β–ͺ πœ–πœ‡ = 0 11

  10. LAGRANGE METHOD How to solve the with two constraints? max 𝑔 𝑦, 𝑧, 𝑨 s.t. ࡞ 𝑕 𝑦, 𝑧, 𝑨 = 𝑑 and β„Ž 𝑦, 𝑧, 𝑨 = 𝑒 Another generalized trick: β–ͺ introduce two Lagrange multipliers πœ‡ and 𝜈 β–ͺ define Lagrangian β„’ 𝑦, 𝑧, 𝑨, πœ‡, 𝜈 β„’ 𝑦, 𝑧, 𝑨, πœ‡, 𝜈 = 𝑔 𝑦, 𝑧, 𝑨 βˆ’ πœ‡ 𝑕 𝑦, 𝑧, 𝑨 βˆ’ 𝑑 βˆ’ 𝜈 β„Ž 𝑦, 𝑧, 𝑨 βˆ’ 𝑒 Find the stationary points β–ͺ by solving five equations: πœ–β„’ πœ–β„’ πœ–β„’ πœ–β„’ πœ–β„’ πœ–π‘¦ = 0 , πœ–π‘§ = 0 , πœ–π‘¨ = 0 , πœ–πœ‡ = 0 , and β–ͺ πœ–πœˆ = 0 12

  11. LAGRANGE METHOD General constraint optimization problem α‰Š max 𝑔 𝐲 s.t. 𝐑 𝐲 = 𝐝 General solution with Lagrangian function β„’ 𝐲, 𝛍 = 𝑔 𝐲 βˆ’ 𝛍 β‹… 𝐑 𝐲 βˆ’ 𝐝 Stationary points of the Lagrangian function are candidate solutions to the optimization problem πœ–β„’ 𝐲,𝝁 πœ–β„’ 𝐲,𝝁 β–ͺ by solving many equations: = 0 and = 0 πœ–π‘¦ 𝑗 πœ–πœ‡ π‘˜ 13

  12. THE LAGRANGE MULTIPLIER We solve the constrained optimization problem by introducing an extra variable πœ‡ (or extra variables 𝛍 ) Setting all partial derivatives of β„’ to 0 , we find the optimal value of 𝑦 and 𝑧 and the optimal value of 𝑔 ... ... but we also find the optimal value for πœ‡ Does it have a meaning? Let’s go back to the simple case α‰Š max 𝑔 𝑦, 𝑧 s.t. 𝑕 𝑦, 𝑧 = 𝑑 14

  13. THE LAGRANGE MULTIPLIER Suppose the constant in the constraint 𝑑 changes β–ͺ recall constraint equation: 𝑕 𝑦, 𝑧 = 𝑑 β–ͺ example, the available budget changes Can we write the problem as a function of 𝑑 ? β–ͺ the optimal values of 𝑦 and 𝑧 depend on 𝑑 β–ͺ so 𝑦 βˆ— = 𝑦 𝑑 and 𝑧 βˆ— = 𝑧 𝑑 β–ͺ and so does the optimal value of 𝑔 = 𝑔 βˆ— 𝑑 β–ͺ can be written as a function of 𝑑 : 𝑔 𝑦 𝑑 , 𝑧 𝑑 Theorem 𝑒𝑔 βˆ— 𝑑 = πœ‡ 𝑒𝑑 15

  14. THE LAGRANGE MULTIPLIER Example Consider the production function optimization problem 𝑔 𝐿, 𝑀 , with 𝐿 capital and 𝑀 labour: α‰Š max 𝑔 𝐿, 𝑀 = 120𝐿𝑀 s.t. 2𝐿 + 5𝑀 = 𝑛 Introduce the Lagrangian β„’ 𝐿, 𝑀, πœ‡ = 120𝐿𝑀 βˆ’ πœ‡ 2𝐿 + 5𝑀 βˆ’ 𝑛 16

  15. THE LAGRANGE MULTIPLIER Example (continued) Solving the first-order conditions (for given 𝑛 ) πœ–β„’ β–ͺ πœ–πΏ = 120𝑀 βˆ’ 2πœ‡ = 0 πœ–β„’ β–ͺ πœ–π‘€ = 120𝐿 βˆ’ 5πœ‡ = 0 πœ–β„’ β–ͺ πœ–πœ‡ = βˆ’2𝐿 βˆ’ 5𝑀 + 𝑛 = 0 yields 1 β–ͺ 𝐿 = 𝐿 𝑛 = 4 𝑛 1 β–ͺ 𝑀 = 𝑀 𝑛 = 10 𝑛 β–ͺ πœ‡ = πœ‡ 𝑛 = 6𝑛 17

  16. THE LAGRANGE MULTIPLIER Example (continued) The maximum value at these maximum points is β–ͺ 𝑔 𝐿 𝑛 , 𝑀 𝑛 = 120𝐿𝑀 = 3𝑛 2 This has been defined above as 𝑔 βˆ— 𝑛 𝑒𝑔 βˆ— 𝑛 Now, consider = 6𝑛 𝑒𝑛 If 𝑛 = 100 and if 𝑛 is increased by Δ𝑛 = 1 to 𝑛 = 101 ; then 𝑔 βˆ— 𝑛 increases approximately by 𝑒 𝑔 βˆ— 𝑛 ቉ Γ— Δ𝑛 = 6 Γ— 100 Γ— 1 = 600 𝑒𝑛 𝑛=100 =πœ‡ 𝑛 18

  17. THE LAGRANGE MULTIPLIER Example (continued) β–ͺ So approximate change of 𝑔 βˆ— is 600 β–ͺ Check approximation with exact result: β–ͺ 𝑔 βˆ— 101 βˆ’ 𝑔 βˆ— 100 = 3 Γ— 101 2 βˆ’ 3 Γ— 100 2 = 603 β–ͺ Good agreement Conclusion The value of a Lagrange multiplier gives the rate of change of the optimum value when the constraint constant is changed by a unit amount 19

  18. EXERCISE 3 In a Lagrange problem, we have πœ‡ = 1.5 and optimum value 𝑔 𝑦 βˆ— , 𝑧 βˆ— = 30 . The budget increases by 4 . What happens? 20

  19. OLD EXAM QUESTION 10 December 2014, Q3c 22

  20. OLD EXAM QUESTION 27 March 2015, Q3d 23

  21. FURTHER STUDY Sydsæter et al. 5/E 14.6 Tutorial exercises week 6 Lagrange method with three variables Lagrange method with two constraints 24

Recommend


More recommend