How to handle missing observations?

Many statistical methods require that no values are missing. In other words, all variables have been measured for all objects. Typically this is not always the case. In metabolomics data sets, some concentrations of metabolites cannot be determined because they are too low to measure. These values are typically referred to as below the detection limit. For typical regression methods to work, these values must be present. In this project you will investigate common procedures to solve this problem. After a short literature investigation, you will identify and implement useful methods for dealing with missing values. Simulation studies are an integral part of the project for demonstrating their usefulness. 


An interest in statistical methods, programming and life sciences (in random order). Familiarity with programming would help (e.g. matlab, r, python).