Optimal sampling designs for multiday and multiperiod panel survey

Abstract

This paper proposes an optimal survey design method for multiday and multiperiod panels that maximizes the statistical power of the parameter of interest. The method addresses balances among sample size, survey duration for each wave, and frequency of observation. Higher-order polynomial changes in the parameter are also addressed, allowing us to calculate optimal sampling designs for nonlinear changes in response to a given policy intervention. After developing the survey design method and showing numerical simulation results, an empirical analysis is conducted using data from the German Mobility Panel, which is an excellent ongoing multiday and multiperiod survey. In the empirical analysis, we identify optimal survey designs for capturing the impacts of policy interventions on trip generation. One of the most important findings in this study is that variation structure in the behavior of interest strongly influences how surveys are designed to maximize statistical power, while the type of policy to be evaluated does not influence it so much. We also point out several important research issues for the future.