creates dummy variables from a data with respect to distinct realizations. The default reference category is the minimal value in each column. Alternatively, categorization can be done by giving a value or the index (rank among the realizations) in a column.
corresp executes Correspondence Analysis which analyses and describes a contingency table cross-tabulations) in terms of a reduced number of dimensions. Correspondence Analysis can be viewed as finding the best simultaneous representation of two sets that comprise the rows and columns of a data mat
CPCp computes the common eigenmatrix, eigenvalues, corresponding standard errors, and estimated population covariance matrices from sample covariances of k groups assuming q common eigenvectors in B; CPCp uses maximum likelihood
CPCprop computes the common eigenmatrix, eigenvalues, correlation coefficients, their standard errors and their estimated population covariance matrices from sample covariances of k groups under the restriction that eigenvalues among groups are linked by a positive constant. Estimation is done usin
factor performs a Factor Analysis for x (principal component, principal axes). For each method you can interactively between two different criteria for the factors. At the end you get a draftman plot of the the chosen factors.
computes sequentially all possible CPC and CPCp models: Beginning with the proportional model, it steps down to the full CPC model and estimates subsequently all possible CPCp models. Additionally to Chi-Square-Statistics, it provides the Akaike (AIC) and Schwarz (SIC) Information Criteria for mode
Calculation of the Kaplan-Meir (product limit) estimator of the hazard rate and the survivor function for a set of durations. The first column of the input is a censorship indicator variable, (equal to zero if the duration is censored, and to one otherwise); the second column is the duration.
provides a summary table (containing: N, Nmiss, min, max, mean, standard error, 1%, 10%, 25%, 50%, 75%, 90%, 99% quantiles) for all columns of a data matrix. Missings values are omitted. An additional vector of name strings can be given to identify the columns by names.