Illuminate the unknown: evaluation of imputation procedures based on the SAVE survey

Abstract

Questions about monetary variables (such as income, wealth or savings) Key words: Imputation methods, Monte-Carlo simulation, imputation evaluation, itemnonresponse, missing data, imputation, retransformation, sample surveys, SAVE JEL classification: C01, C81, C49 1 I would like to thank Michela Coppola (MEA) for very helpful support, as well as Axel Börsch-Supan (MEA), Daniel Schunk (Johannes Gutenberg University Mainz), Armin Rick (University of Chicago), Mathias Sommer, Hendrik Jürges (Universität Wuppertal), Uli Schlieper (MEA), Cornelius Schmidt (Université de Lausanne) for their helpful comments. Alexander Aldinger, Helen Grupp, Dörte Heger, Corinna Kartzke, Julius Nick, Michael Schardt, Stefanie Schulz, and Marius Wolter provided excellent research assistance. I am grateful to the seminar participants at MEA for many useful comments. MEA is a research institute which is funded by two thirds through public third-party means for which I am very grateful. Furthermore, I would like to thank the state of Baden-Württemberg and the German Insurance Association for financing the basic funding of MEA. I am particularly grateful to the German Research Foundation (Deutsche Forschungsgemeinschaft) for financing the SAVE survey.

Skilled But Unaware of It: Occurrence and Potential Long-Term Effects of Females' Financial Underconfidence

“…This should increase the efficiency of estimates -based on the larger number of observations -and reduce the non-response bias. 7 Missing observations in our dataset are therefore imputed using an iterative multiple imputation procedure based on a MarkovChain Monte-Carlo method (Schunk, 2008;Ziegelmeyer, 2013). We use five multiply imputed data sets for our analysis and results are derived by using Rubin's rule (Rubin, 1987(Rubin, , 1996.…”

Actual and Perceived Financial Sophistication and Wealth Accumulation: The Role of Education and Gender

“…We observe that respondents who experienced adverse crisis effects show higher financial wealth than those who were not negatively affected by the crisis (Panel A). As Bucher-Koenen and Ziegelmeyer (2013) clarify, this surprising observation is explained by the fact that the respondents whose wealth was reduced in the crisis were active on financial markets and profited from the markets' resurgence. Typically, these are individuals with high financial literacy.…”

Actual and Perceived Financial Sophistication and Wealth Accumulation: The Role of Education and Gender

“…This should increase the efficiency of estimates -based on the larger number of observations -and reduce the item non-response bias. 3 Missing observations in our dataset are therefore imputed using an iterative multiple imputation procedure based on a MarkovChain Monte-Carlo method (Schunk, 2008;Ziegelmeyer, 2013). We use five multiply imputed data sets for our analysis and results are derived by using Rubin's rule (Rubin, 1987(Rubin, , 1996.…”