
The synthetic difference-in-differences (SDiD) method can be adapted to estimate causal effects using repeated cross-sectional data by incorporating a third type of weight that accounts for the varying number of observations in each group-period.


coremsg

adapting-synthetic-difference-in-differences-for-repeated-cross-sectional-data


Adapting Synthetic Difference-in-Differences for Repeated Cross-Sectional Data


title_rewrite


The author demonstrates that even small measurement errors in network diffusion models can lead to significant forecasting inaccuracies, impacting the estimation of diffusion counts and the sensitivity to initial seed identification.


non-robustness-of-diffusion-estimates-on-networks-with-measurement-error


Non-Robustness of Diffusion Estimates on Networks with Measurement Error
