Consider a simple pooled regression model: yit=β0+β1xit+vit where vit=μi+ϵit and Cov(xit,μi)=0. Here, μi captures the unknown individual specific effects and ϵit is the idiosyncratic error uncorrelated with both xit and μi. If the parameters of this model are estimated using the ordinary least squares (OLS) method, then the estimated slope coefficient will be
Select all that apply.
Answer and worked solution
Checking your account…
Plain-text version
Consider a simple pooled regression model: yᵢₜ = β₀+β₁xᵢₜ+vᵢₜ where vᵢₜ=μᵢ+ϵᵢₜ and Cov(xᵢₜ,μᵢ)≠ 0. Here, μᵢ captures the unknown individual specific effects and ϵᵢₜ is the idiosyncratic error uncorrelated with both xᵢₜ and μᵢ. If the parameters of this model are estimated using the ordinary least squares (OLS) method, then the estimated slope coefficient will be (A) biased (B) inconsistent (C) unbiased but consistent (D) unbiased but efficient
Also written as: Consider a simple pooled regression model: y_(it) = beta_0+beta_1x_(it)+v_(it) where v_(it)=mu_i+epsilon_(it) and Cov(x_(it),mu_i)!= 0. Here, mu_i captures the unknown individual specific effects and epsilon_(it) is the idiosyncratic error uncorrelated with both x_(it) and mu_i. If the parameters of this model are estimated using the ordinary least squares (OLS) method, then the estimated slope coefficient will be
Data types and panel data has come up 5 times in 4 exams since 2015.
What this question tests
This question is on Panel data: fixed and random effects, part of Data types and panel data in Econometrics.
Models for data on the same units over time, and how to choose between them.
TestStop has 2 past-paper questions on this subtopic.