Imputation-Codes

Missing data are a common problem in many fields of human endeavor ranging from social sciences to economics and from political research to entertainment industry. The objective of this thesis is to study and apply missing data algortihms for treatment of missing data which encountered on bank marketing data sets. Six missing data imputation algortihms are analyzed which are mean, hot-deck, k-nearest neighbor, regression, expectation maximization and multiple imputation method. By imputation of missing data, the accuracy of clustering and association with regard to aspect of the data can be increased. This thesis also explores the advantages and disadvantages of each used imputation method. Similarities and differences between each imputation method is discussed, as well.