Model Assessment Sample Clauses

Model Assessment. It is important to conduct diagnostic measurements to see whether the model is an xxx- xxxxx fit with justified assumptions. For example, standard linear regression requires the normality, independence, linearity, and homogeneity of variance [Xxxxxx and Xxxxx, 2009]. Similar to classical approach, we define a Bayesian residual as ri = yi −E(Yi|z), i = 1, ..., n, where z = (z1, ..., zm) is the fitting sample and y = (y1, ..., yn) is the validation sample. Cross-validatory The limited number of observations in cancer mortality data per age group pushes us to consider the cross-validatory (or ”leave one out”) approach [Xxxxxx and Xxxxx, 2009], where the estimated value for xx is computed conditional on all the data except yi. In other words, we calculate the Bayesian residual as ri = yi − E(Yi|y(i)), i = 1, ..., n, where y(i) = (y1, ..., y(i—1), y(i+1), ..., yn)r. And the standardized residual is defined as x x xx − E(Xx|y(i)) di = .