The study aims to apply the Multiple Imputation (MI) method in case of missing observation in the quantitative data and to determine the covariance structure between the repeated measures. In estimating the missing observations, missing observations were assumed to be Missing at Random (MAR) and MCMC (Markov Chain Monte Carlo) technique and multiple imputation method were applied. To that end, live-weight data with missing observation and quantitative structure was used. Time factor was included as a continuous variable into the model that was formed to evaluate the live-weight data and random intercept and slope model were created. Compound Symetry (CS), Heterogenous Compound Symetry (CSH), Unstructured (UN), First order Autoregressive (AR (1)), Heterogenous first order Autoregressive (ARH (1)), Toeplitz (TOEP) and Heterogenous Toeplitz (TOEPH) structures were used to determine the covariance structure between repeated measurements in the data sets that have missing observations and missing observations which were estimated. Consequently, CS, AR (1) and TOEP structures were assumed to be the best model according to the AIC and BIC cohesion goodness of fit in modeling covariance matrix structure regarding the variable in the model established on the repeated measure data handled in both cases. UN, CSH, TOEPH and ARH (1) were found to be the worst model with heterogeneous covariance structure.