What time use surveys can (and cannot) tell us about labor supply
研究了如何利用美国时间使用调查（ATUS）的日度数据来估计周劳动供给弹性，提出了一种改进的插补估计量，并发现其估计结果与当前人口调查（CPS）有显著差异，尤其对已婚女性。
Summary The American Time Use Survey (ATUS) accurately measures hours worked on a single day. We propose several estimators of elasticities of weekly labor supply in a linear regression model, despite certain impossibility results due to the time specific feature of the ATUS. We recommend the impute estimator , a simple modification of the standard two stage least squares estimator, that imputes the dependent variable using daily subsamples, based on our careful investigation of asymptotic and finite sample properties of the estimators under the potential outcome framework. We apply the impute estimator to the ATUS and find substantially different elasticity estimates from the Current Population Survey, especially for married women.