Surveys are a widely used tool to answer socio-economic research question across disciplines. However, data collection can face certain problems such as nonresponding units. For household and population surveys, a large body of literature about the effects of nonresponse exist but only less is known in case of business surveys. This thesis deals with the missing values in the Ifo Business Survey which is conducted in similar form in nearly all OECD countries. The most prominent result of this survey is the Ifo Business Climate Index, a business cycle indicator for the German economy. This indicator is highly observed by entrepreneurs, analysts, politicians, journalists, academics and the general public. The results of this thesis show that business cycle indicators based on this type of questioning are very stable towards any kind of non-random missing data processes. This is shown by simulation studies as well as an estimation of the missing values. In particular, the missing values do not lead to a significant reduction in forecasting performance.
Business Survey Ifo Business Climate Imputation Missing data Panel survey Unit nonresponse