UC Davis Agricultural and Resource Economics

Tamma Carleton, University of California, Santa Barbara

Parameter recovery using remotely sensed variables

Date and Location

Wednesday, November 8, 2023, 1:10 PM - 2:30 PM
ARE Library Conference Room, 4101 Social Sciences and Humanities


Remotely sensed measurements and other machine learning predictions are increasingly used in place of direct observations in empirical analyses. Errors in such measures may bias parameter estimation, but it remains unclear how large such biases are or how to correct for them. We use a large dataset of diverse remotely sensed variables paired with ground truth observations to establish a set of stylized facts regarding parameter recovery with remotely sensed observations. We show that using remotely sensed variables without correction leads to substantial biases in point estimates and standard errors that are accentuated when these variables are used as regressors. For example, more than three-quarters of the 95% confidence intervals we estimate using remotely sensed measurements do not contain the true coefficient of interest. Additionally, we decompose these biases to establish that both classical and nonrandom measurement error are responsible for parameter biases. We then demonstrate that multiple imputation, a standard and easily implementable statistical imputation technique generally applied to missing data problems, effectively reduces bias and improves statistical coverage in both cross-sectional and panel data designs.

Subscribe to Upcoming Seminars

Click here to receive weekly notice of all upcoming seminars via email.

Individual seminar and workshop announcements are circulated via separate mailing lists. To subscribe, send an email to the relevant list:

Contact Us

2116 Social Sciences and Humanities
University of California, Davis
One Shields Avenue
Davis, CA 95616

Main Office: 530-752-1515
Student Advising Services: 530-754-9536
DeLoach Conference Room: 530-752-2916
Main Conference Room: 530-754-1850