EN
The purpose of this paper is to present dynamic approach to selection of efficient portfolios using a forecasts of variances and covariances from the multivariate GARCH models. Evaluation of efficiency for different methods of asset allocation is also performed for stocks from the WSE. Twelve specifications of the multivariate GARCH models, the univariate GARCH model and six other covariance matrix estimation methods are used. Taking into consideration time varying variances and covariances of stock returns in portfolio selections increases, with some exceptions, efficiency of asset allocation process. Simple specifications of them ultivariate GARCH models, w hich parameters are estimated in one stage, are the best perform ing models. From economic point of view, the differences between the models are not significant, with exception of the factor and orthogonal models. RiskMetrics methodology commonly used by practitioners does not give good results for constructions of efficient portfolios.