{[ promptMessage ]}

Bookmark it

{[ promptMessage ]}

# 140_problems7_answers - ECON 140 Fall 2008 11/6 Alex...

This preview shows pages 1–3. Sign up to view the full content.

ECON 140, Fall 2008 - 11/6 Alex Rothenberg Practice Problems: Serial Correlation and Panel Data Problem 1 Suppose we have the following autoregressive model for the error term (the unobservables) in a regression equation: ε t = ρ ε t - 1 + u t (1) u t iid N (0 , σ 2 u ) (2) This question asks you to investigate some of the properties of this process. 1. Fixing the value of ε 0 = 0, write out the first few terms of the sequence { ε t } in terms of the u t ’s. (Do this for t = 0 , 1 , 2 , 3 , 4) ANSWER: We write out the terms of the sequence as follows (using recursive substitution): ε 1 = ρ ε 0 + u 1 = u 1 ε 2 = ρ ε 1 + u 1 = ρu 1 + u 2 ε 3 = ρ ε 2 + u 3 = ρ 2 u 1 + ρu 2 + u 3 ε 4 = ρ ε 3 + u 4 = ρ 3 u 1 + ρ 2 u 2 + ρu 3 + u 4 2. Assuming E [ ε 0 ] = 0, what is the expected value of each term? ANSWER: Because each of the ε t ’s is just a linear combination of the u t ’s, which are mean zero random variables, each ε t will also be mean zero. For example: E [ ε 3 ] = ρ 2 E [ u 1 ] + ρ E [ u 2 ] + E [ u 3 ] = ( ρ 2 × 0) + ( ρ × 0) + 0 = 0 3. Assuming that Var( ε t ) = σ 2 for all t (homoskedasticity), express σ 2 as a function of σ 2 u . ANSWER: Returning to the recursion equation, we can apply the variance operator: Var( ε t ) = Var( ρ ε t - 1 + u t ) = ρ 2 Var( ε t - 1 ) + Var( u t ) 1

This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document
ECON 140, Fall 2008 - 11/6 Alex Rothenberg Where the second line makes use of the independence between u t and ε t - 1 . Invoking the homoskedasticity assumption, we have: σ 2 = ρ 2 σ 2 + σ 2 u σ 2 = σ 2 u 1 - ρ 2 Note that this trick of solving for σ 2 and dividing by 1 - ρ 2 is not possible if ρ = 1 or ρ = - 1 . In either case, the process (1) is a random walk, and dealing with it is tricky business. We will discuss more about stationarity later in the course (the
This is the end of the preview. Sign up to access the rest of the document.

{[ snackBarMessage ]}