Week 10 Homework Feedback: Matteo Pugliese

Overall Assessment

Grade: ✅ (Pass)

Correct results, but different method. You solved the model using Value Function Iteration (VFI) instead of the requested Endogenous Grid Method (EGM) or Coleman policy iteration. While VFI is a valid method and you implemented it correctly to get the right economic results, the assignment aimed to practice EGM/Coleman which are faster and more accurate for Euler equation errors.

Task-by-Task Check

Task 1: Quantifying Precautionary Savings

  1. Parameter & Grid Setup: ✅ Correct.
  2. Solution Method: ⚠️ Used VFI (max over grid) instead of EGM.
  3. Deterministic Case: ✅ Solved.
  4. Stochastic Case: ✅ Solved.
  5. Simulation: ✅ Simulation logic is correct.
  6. Comparison: ✅ Premium calculated.

Task 2: The Role of Risk Aversion

  1. Sigma Variation: ✅ Implemented.
  2. Solving Sigma Cases: ✅ Solved.
  3. Policy Plot: ✅ Plotted.
  4. Simulation Comparison: ✅ Calculated.
  5. Interpretation: ✅ Correct.

Technical Implementation

  • VFI vs EGM: Your solve_model function iterates on V and finds the policy by maximizing u + beta * EV. This is VFI. It works, but it’s generally slower and limited by the grid resolution compared to EGM/Coleman which use the first-order conditions directly.
  • Correctness: Despite using a different method, your implementation of VFI is correct and robust.

Visual Output

  • Figures are clear and correct.

Suggestions for Improvement

  1. Method Selection: For future reference, EGM is preferred for this type of continuous-choice problem because it avoids the discrete maximization step on the grid, allowing for smoother policy functions and faster convergence.

Summary

10/11 tasks correct (method choice variation). Correct work.