Abstarct: In the most applications, researchers are imterested in finding out a relationbetweenaresponseandsomecovariates. Forsuchaim, aregressionmodel is usually implemented. In many practical situation, the response variable is continuous and restricted to the interval (0,1); e.g. proportions, rates and percentages. For examples: in econometrics, peaple are interested in finding out the relation between growth rate, unemployment rate, and GDP percentage with several other economic variables. To modeling such responses, the common models are logistic and probit regressions. However, the proportions and rates are usually concentrated in a specific subinterval. In other words, such responses are skewed. Therefore, the logistic and probit models are not appropriate for modeling these responses. Regarding to these constraints, the appropriate and efficient model for this kind of data is the novel and beta regression model. In this thesis, we first define the proposed model and then estimate the parameters by maximum likelihood approach. We establish the asymptotic properties of the extracted estimators as well. The efficiency of beta regression model and its asymptotic properties are explored by a few simulation studies. Finally, we apply the proposed model to analyze two examples; first example is on gas percent output from crude oil, and the second is on concrete resistivity.