13+14 Ramsey Flashcards
Which constraints is this optimization problem subject to?
- transition equation
- terminal condition
- initial condition
What are the components of this optimization problem? What are their roles?
e<sup>-ρt</sup> = discount factor in continuous time cₜ = consumption = control variable, which we maximize kₜ = net assets, capital = state variable (as all other variables that are not control) t = time
What is the transition equation?
differential equation in k that shows how the choice of control variable translates into a pattern of movement for the state variable(s)
What is the terminal condition?
chosen discrete value of the state variable at the end of the planning horizon
What is the initial condition?
initially, the state variable must be non-negative
Following our standard Lagrange approach, we would maximize the Lagrangian w.r.t ct and kt. Why can we not do that?
We don’t know what the partial derivative of k̇ₜ with respect to kₜ is
What is the cookbook recipe for applying the Hamiltonian?
- identify control + state variables
- define the Hamiltonian: (function) + λₜ (transition equation)
- take the partial derivatives w.r.t. control and state variables and set the one for the control variable equal to zero, the state variable equal to -dot λt
- enforce the transversality condition
What do the different parts of the Hamiltonian mean?
(1) the first part is the direct contribution to the objective of the control variable
(2) the second part shows how the choice of the control variable affects the state variable’s evolution. The value of the change in the state variable is captured by λg() = the rate of change weighted by its shadow value (Lagrangian multiplier)
What is the transversality condition?
It describes what happens at the end of the planning horizon (T): if the state variable left at T is positive, then the shadow value (= λ_T) must be zero - otherwise it is not optimal not to use the remaining state variable any more. Vice versa, if there is not state variable left, it implies that the shadow value is positive.
What are phase diagrams?
- Graphical device to solve a system of equations
- Allows us to visualize the dynamics of the system
- Objective: translate the differential equations into a system of arrows that describe the qualitative behavior of the economy over time
In a phase diagram, what is the locus?
- It is the set of points for which ẏ₁(t) = 0
- It is called the y₁ = 0 schedule
What are the steps to construct a Phase Diagram?
- Plot the locus of the points for which ẏ₁(t) = 0
- Analyze the dynamics of y₁ in the two regions generated by ẏ₁ = 0, i.e. where y₁ is positive & where y₁ is negative
- Repeat 1.+2. for y₂
- Join the two graphsand identify the steady state
Which steady states are there in a system (phase diagram)?
- unstable steady state
- stable steady state
- saddle-path stable steady state
What is an unstable steady state (phase diagram)?
A steady state, where if you are only a little bit off, the dynamics of the system will take you away from the steady state
What is an stable steady state (phase diagram)?
No matter where in the system you start, the system will take you back to the steady state (at the origin)