LN+12+Overview+of+Data+Issues+and+Forecasting

LN+12+Overview+of+Data+Issues+and+Forecasting - Empirical...

Info iconThis preview shows pages 1–4. Sign up to view the full content.

View Full Document Right Arrow Icon
Empirical Methods II (API-202) Kennedy School of Government Harvard University 1 Lecture Notes 12 Overview of Data Issues and Forecasting Today : An overview of how to think about and deal with: o Data issues that you might encounter when working with data: Missing data (Sample selection) Quality of the variables (Measurement error or errors-in-variables) Outliers o Forecasting Are we always interested in causality? How do we select a regression model to make a prediction? (i.e., which variables to include)?
Background image of page 1

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

View Full DocumentRight Arrow Icon
Empirical Methods II (API-202) Kennedy School of Government Harvard University 2 I. OVERVIEW OF DATA ISSUES Missing data (Sample selection) Key question: Does the missing data follow a particular pattern? i) Are more data points missing for higher/lower values of X? ii) Are more data points missing for higher/lower values of Y? iii) Are data missing just randomly? Sample selection will only be a problem for causal identification (internal validly threat) in case ii. Only on this case, the condition E(u|X)=0 does not hold due to sample selection. 1 For example: regressing wages on experience. wages = 0 ˆ 1 ˆ experience + ˆ Common approaches to sample selection: 2SLS (next class), Heckman correction (advanced topic) 1 See Wooldridge page 606 for more details.
Background image of page 2
Empirical Methods II (API-202) Kennedy School of Government Harvard University 3 Quality of the variables (Measurement error) Our model of interest: Y = 0 ˆ 1 ˆ X + ˆ But we can only measure Y and X with error: Y * and X * o Y * = Y+e o X * = X+g o e and g represent the measurement errors in each variable QUESTION : Do you think that measurement error is a problem for causal interpretation? Does
Background image of page 3

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

View Full DocumentRight Arrow Icon
Image of page 4
This is the end of the preview. Sign up to access the rest of the document.

This note was uploaded on 04/12/2009 for the course HKS API202A taught by Professor Levy during the Spring '09 term at Harvard.

Page1 / 7

LN+12+Overview+of+Data+Issues+and+Forecasting - Empirical...

This preview shows document pages 1 - 4. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online