Large-scale complex surveys typically contain a large number of variables measured on an even larger number of respondents. Missing data is a common problem in such surveys. Since usually most of the variables in a survey are categorical, multiple imputation requires robust methods for modelling...

In the first part of the dissertation, we derive two methods for responders analysis in longitudinal data with random missing data. Often a binary variable is generated by dichotomizing an underlying continuous variable measured at a specific point in time according to a prespecified threshold...

Monitoring soil quality provides indispensable inputs for effective policy advice, but very few poorer countries can implement high-quality surveys on soil. We offer an alternative, low-cost imputation-based approach to generating various soil quality indicators. The estimation results validate...

Monitoring soil quality provides indispensable inputs for effective policy advice, but very few poorer countries can implement high-quality surveys on soil. We offer an alternative, low-cost imputation-based approach to generating various soil quality indicators. The estimation results validate...

The results of a national fear of crime survey are compared with results following the use of different nonresponse correction procedures. We compared naive estimates, weighted estimates, estimates after a thorough nonresponse follow-up and estimates after multiple imputation. A strong...

Smallholder farming dominates agriculture in poorer countries. Yet, traditional recall-based surveys on smallholder farming in these countries face challenges with seasonal variations, high survey costs, poor record-keeping, and technical capacity constraints resulting in significant recall...

Smallholder farming dominates agriculture in poorer countries. Yet, traditional recall-based surveys on smallholder farming in these countries face challenges with seasonal variations, high survey costs, poor record-keeping, and technical capacity constraints resulting in significant recall...

Abstract Data mining and machine learning techniques such as classification and regression trees (CART) represent a promising alternative to conventional logistic regression for propensity score estimation. Whereas incomplete data preclude the fitting of a logistic regression on all subjects,...

We estimate how motorists value their time savings and characterize the degree of heterogeneity in these values by observable traits. We obtain these estimates by analyzing the choices that commuters make in a real market situation, where they are offered a free-flow alternative to congested...