Ch. 7: Rules of Differentiation and Their Use in Comparative Statistics

The central problem of comparative-static analysis, that of finding a rate of change, can be identified with the problem of finding the derivative of some function, provided only an infinitesimal change in x is being considered. 

7.1: Rules of Differentiation for a Function of One Variable

7.2: Rules of Differentiation Involving Two or More Functions of the Same Variable

7.3: Rules of Differentiation Involving Functions of Different Variables

Ch. 8: Comparative-Static Analysis of General-Function Models

Partial derivatives enables us to handle problems in which the equilibrium solution of a model can be explicitly stated in reduced form. Recall that the definition of the partial derivative requires the absence of any functional relationship between the independent variables (no co-linearity. 

8.1: Differentials

By definition, the derivative is the limit of a difference quotient:
\(\frac{dy}{dx} = f'(x) = lim_{\Delta x \rightarrow 0} \frac{\Delta y}{\Delta x}\)
The derivative can also be reinterpreted as the factor of proportionality between the two finite changes \(dy \) and \(dx\)
Differentials and Point Elasticity: 
Given a demand function \(Q = f(P)\), it's point elasticity of demand at certain quantities would be: \(\epsilon _d = \frac{\frac{dQ}{Q}}{\frac{dP}{P}} = \frac{dQ}{Q} \frac{dP}{P}\)
The point elasticity of demand is seen to be the ratio of the marginal function to the average function of the demand function. 

Ch. 9: Optimization: A Special Variety of Equilibrium Analysis

In optimization, the first order of business is to delineate an objective function in which the dependent variable is the object of maximization/minimization, and the independent variables are those that can be picked and changed, thus choice variables. For example, a firm may seek to maximize profit, that is, the difference between total revenue and total cost:
\(\pi (Q) = R(Q) - C(Q)\)

9.4 First Derivative Test

  1. Find the derivative function
  2. Get the critical values of x that satisfy the condition
  3. Verify with the second derivative whether it is a relative maximum or minimum
The slope of zero is the first, necessary, condition. The determined sign of the second derivative is the second, sufficient, condition. 

Ch. 11: The Case of More than One Variable

Now we shall discuss an objective function with two choice variables. 

______________________

Class Notes

Test-like question
Annuity-like formula with growing payment, growing at rate g:
Max \(\int ^T _0 \frac{C_t ^{1 - \theta}}{1 - \theta}dt\) where \(\dot K = rK - c\) (fixed rate of return) and \(K(0) = \bar K\)
\(K_0 = \frac{C_0}{r - g} (1 - e^{-(r - g)T})\) (first period consumption divided by the real interest rate minus growth rate, multiplied one minus the rate of growth)
\(C_t = C_0 e^{gt}\) (for any time, consumption will equal starting consumption growing at rate g)
Solve for \(C_0\), taking \(K_0, r, g, T\) as given. The stock variable is capital, while the flow is the consumption. 
Quiz Explaination:
Question 1: \(\frac{dx}{dt} = 3 - 2x_t\)
\(x(0) = 0\)
Just divide solve for \(\frac{dx}{dt} = 0\)
 Question 2: \(V = \int^1 _0 [x_t - \frac{10}{2} u_t^2]dt\)
\(H = x - 5u^2 + \lambda u\)
\(\frac{\partial H}{\partial u} = 0 = 10u + \lambda \rightarrow - \lambda = 10u\)
\(\frac{\partial H}{\partial x} = 1 = - \lambda \rightarrow \lambda = -t + c_1\)
\(\lambda(1) = 0\)
\(\therefore 10 u = t + c_1 \Longrightarrow u = \frac{t}{10}\)
Taylor Series
A widely-used tool in computational work. Sometimes used to create approximations to highly nonlinear models. Fact: any function can be approximated to an arbitrarily precise degree by a polynomial.
EX.: \(f(x) = a_0 + a_1x + a_2x...\)
Given this, let's find some derivatives:
\(f'(x) = a_1 + 2a_2x + 3a_3x^2 ...\)
\(f''(x) = 1 \cdot 2 + a_2x + 2 \cdot 3 a_3x...\)
Look at \(x = 0\)
\(f(0) = a_0\)
At \(x = 0\), I know that
\(f'(x) = a_1\)
\(f''(x) = 1 \cdot 2 \cdot a_2\)
But look around at X near 0