1 2 al algorithm allows analyses analysis applications applied approach arbitrary association available basic bayesian binary book bootstrap c calculate calculation carlo censored chain class classes classification cluster clustering code collection common components computation computational compute computing conditional confidence control correlation count covariates create currently curves data database datasets density described design designed detection different discrete display distance distribution either engineering environment error estimate estimating estimation estimator et etc exact examples experiments features file finance financial first fit fitting framework function functionality gaussian gene general generalized genetic graph graphical graphics group gui hazard hierarchical if implementation implemented implements include included including independent inference information interface intervals its kernel large level library likelihood linear local logistic main manipulating map markov matrices matrix maximum may mean measures method microarray missing mixture model modeling modelling monte most multiple multivariate network nonlinear nonparametric normal number object observations order output package parameter parametric perform plot plotting point population possible power probability problems procedure process processes program programming proportional provide provided quantitative r random regression related response results risk robust routines s sample sampling selection series set simple simulation single smoothing software spatial specified splus squares standard statistical statistics structure support survival system teaching test testing theory through time tools trees univariate useful user uses using utilities utility value variable variance various vector version very visualization wavelet way weighted work written