[gov3009-l] Blackwell on "Multiple Overimputation: A Unified Approach to Measurement Error and Missing Data"

Matt Blackwell mblackwell at iq.harvard.edu
Mon Sep 13 09:26:42 EDT 2010


Hi all,

We hope you can join us for the Applied Statistics Workshop this
Wednesday, September 15th, when I will have the dubious honor of
hosting myself, Matthew Blackwell, and presenting joint work with
James Honaker and Gary King. Details of the talk and an abstract are
below. The workshop begins at 12 noon and we will serve a light lunch.
Hope to see you there!

"Multiple Overimputation: A Unified Approach to Measurement Error and
Missing Data"
Matthew Blackwell (with James Honaker and Gary King)
Governemnt
September 15th, 2010, 12 noon
K354 CGIS Knafel (1737 Cambridge St)

Abstract:
Social scientists typically devote considerable effort to reducing
measurement error during data collection only to ignore the issue
during data analysis. Although many statistical methods have been
proposed for reducing measurement error-induced biases, few have been
widely used because of implausible assumptions, high levels of model
dependence, difficult computation, or inapplicability with multiple
mismeasured variables. We develop an easy-to-use alternative that
generalizes the popular multiple imputation (MI) framework so that it
treats missing data problems as a special case of extreme measurement
error and corrects for both. Like MI, the proposed "multiple
overimputation" (MO) framework is a simple two-step procedure. First,
multiple (~5) completed copies of the dataset are created where cells
measured without error are held constant, those missing are imputed
from the distribution of predicted values, and cells (or entire
variables) with measurement error are "overimputed," that is imputed
from a predictive distribution with observation-level priors defined
by the mismeasured values and available external information, if any.
In the second step, analysts can then run whatever statistical method
they would have run on each of the overimputed datasets as if there
had been no missingness or measurement error; the results are then
combined via a simple procedure. We also offer open source software
that implements all the methods described herein.

You can find a copy of the paper here:
http://gking.harvard.edu/files/abs/measure-abs.shtml

Cheers,
matt.


~~~~~~~~~~~
Matthew Blackwell
PhD Candidate
Institute for Quantitative Social Science
Department of Government
Harvard University
url: http://people.fas.harvard.edu/~blackwel/


More information about the gov3009-l mailing list