Tensile characteristics of yarn have a great effect on performance of textile products. In this study, tensile characteristics (i.e. breaking force and elongation properties) of chenille yarns are investigated using theoretical and statistical approaches. The aim of this investigation is to explain the effect of selected production parameters and yarn components on tensile properties of chenille yarn and to develop some practical models to predict the tensile properties before production. In theoretical approach, a mathematical prediction model was developed by using geometrical methods for breaking force of chenille yarn. 27 different chenille yarn samples were produced according to an experimental design which contains different levels of effective factors. In statistical approach, the data sets were divided into two parts of 20 and 7 sets. First data group was used in stepwise regression analysis and second data group was used for verification. Finally, the mathematical and statistical models were evaluated and compared their performances in terms of estimating tensile properties. As a result, it has proved that breaking force of chenille yarn can be estimated by theoretical and statistical methods. The significant effective factors on breaking force of chenille yarn are count of binder yarn and twist level of chenille yarn. It was not obtained a strong statistical model for elongation of chenille yarn but the model was quite succesfull in verification.