Classification and Regression Tree (CART) Techniques

In certain research studies development of a reliable decision rule, which can be used to classify new observations into some predefined categories, plays an important role. The existing traditional statistical methods are inappropriate to use in certain specific situations, or of limited utility, in addressing these types of classification problems. There are a number of reasons for these difficulties. First, there are generally many possible “predictor” variables which makes the task of variable selection difficult. Traditional statistical methods are poorly suited for this sort of multiple comparisons. Second, the predictor variables are rarely nicely distributed. Many variables (in agriculture and other real life situations) are not normally distributed and different groups of subjects may have markedly different degrees of variation or variance. Third, complex interactions or patterns may exist in the data. For example, the value of one variable (e.g., age) may substantially affect the importance of another variable (e.g., weight). These types of interactions are generally difficult to model and virtually impossible to model when the number of interactions and variables becomes substantial. Fourth, the results of traditional methods may be difficult to use. For example, a multivariate logistic regression model yields a probability for different classes of the dependent variable, which can be calculated using the regression coefficients and the values of the explanatory variable. But practitioners generally do not think in terms of probability but, rather in terms of categories, such as “presence” versus “absence.” Regardless of the statistical methodology being used, the creation of a decision rule requires a relatively large dataset.