Econometrics of Network and Panel data
Saturday, Jan. 5, 2019 10:15 AM - 12:15 PM
- Chair: Eric Auerbach, Northwestern University
Misclassification and the Hidden Silent Rivalry
AbstractThe interaction of economic agents is one of the most important elements in economic analyses. While most empirical studies investigate peer effects on objective final achievements, peer effects on subjective outcomes are inherently difficult to identify and estimate because these variables are prone to measurement errors. In particular, peer effects on students' attitudes towards learning are believed to have a significant impact on their achievements, while we found the presence of misclassification errors in students' self-reported attitudes. We develop a binary choice model with misclassification and social interactions and use a recently developed technique of measurement error models to correct misreporting errors for estimating the peer effects on attitude. Our estimates suggest that a significant proportion of students overreport their attitudes towards learning and that peer effects are not only significant, but also much larger than estimates ignoring the misreporting errors. Our method may be generalized to the identification and estimation of peer effects with imperfect data information.
Identification of an Interactive Panel Data Model with Fixed Effects
AbstractPanel data are often used to allow for unobserved individual heterogeneity in econometric models. In this paper we discuss identification of a panel data model with fixed effects in which the fixed effects are allowed to interact with covariates. We also discuss identification of a more general model which also allows the lagged dependent variables to have certain types of explicit ceteris paribus effects on the current period dependent variable.
Identification and Estimation of a Partially Linear Regression Model using Network Data
AbstractI study a regression model in which one covariate is an unknown function of a latent driver of link formation in a network. Rather than specify and fit a parametric network formation model, I introduce a new method based on matching pairs of agents with similar columns of the squared adjacency matrix, the ijth entry of which contains the number of other agents linked to both agents i and j. The intuition behind this approach is that for a large class of network formation models the columns of this matrix characterize all of the identifiable information about individual linking behavior. In the paper, I first describe the model and formalize this intuition. I then introduce estimators for the parameters of the regression model and characterize their large sample properties.
- C2 - Single Equation Models; Single Variables
- D8 - Information, Knowledge, and Uncertainty