Fig. 2From: Evaluation of data imputation strategies in complex, deeply-phenotyped data sets: the case of the EU-AIMS Longitudinal European Autism ProjectEach row presents information about one of the fifteen subsets. The first column (left) presents missing data as blue dots with individuals presented in the x-axis and number of clinical measures in the y-axis. The second and third columns present the percentage of shared missing variables per pair of individuals, and the percentage of missing individuals per pair of behavioural measures respectively, with darker colours coding an increased percentage. The fourth and fifth columns present histograms showing the number of individuals missing a number of variables, and the number of variables being missed by a number of individuals. The sixth column present the correlation structure inside each of the subsets i.e. diagonal subsquares of Fig. 1Back to article page