In diagnostic testing,model specification tests play an important role. The specification of a good model is an art. Over specification yields unbiased estimates of regression coefficients,but larger variance; under specification yields biased estimates of regression coefficients and understates the variance of these estimates. In case of the left out variables or because of inclusion of variable not specified by the truth; the case of irrelevant variables. Mis-specification is usually interpreted as a case of omitted variables,and many researchers,concerned only with the bias resulting from it,the specification bias. Researchers seldom pay attention to the other aspects of mis-specification of the model. In view of the importance of these aspects of misspecification in empirical research,some major results of misspecification error tests are considered in this Article. In this research article,three types of new tests for the mis-specification of the linear regression model have been developed by using various types of residuals. © 2016,International Journal of Pharmacy and Technology. All rights reserved.