Nearest Neighbor analysis – Use for prediction (with a specified outcome) or for classification (with no outcome specified) specify the distance metric used to measure the similarity of cases and control whether missing values or categorical variables are treated as valid values. Ordinal regression-PLUM – Choose from seven options to control the iterative algorithm used for estimation, to specify numerical tolerance for checking singularity, and to customize output five link functions can be used to specify the model. Linear Regression – Choose from six methods: backwards elimination, forced entry, forced removal, forward entry, forward stepwise selection and R2 change/test of significance produces numerous descriptive and equation statistics. Discriminant – Offers a choice of variable selection methods, statistics at each step and in a final summary output is displayed at each step and/or in final form.
Compare means – Choose whether to use harmonic or geometric means test linearity compare via independent sample statistics, paired sample statistics or one-sample t test.Descriptive ratio statistics – Coefficient of dispersion, coefficient of variation, price-related differential and average absolute deviance.Descriptives – Central tendency, dispersion, distribution and Z scores.Frequencies – Counts, percentages, valid and cumulative percentages central tendency, dispersion, distribution and percentile values.
Crosstabulations – Counts, percentages, residuals, marginals, tests of independence, test of linear association, measure of linear association, ordinal data measures, nominal by interval measures, measure of agreement, relative risk estimates for case control and cohort studies.