PC-MDS: MULTIDIMENSIONAL STATISTICS FOR THE PC Thank you for requesting information about PC-MDS, the world's most widely adopted specialty package for multi-dimensional scaling and multivariate analysis. The PC-MDS programs are based on industry standard programs developed by Bell Laboratories and BMD (Developed under a federal research grant). PC-MDS programs have been developed for production work. Each PC-MDS program operates as an independent unit. For example, Notebook computer users may copy several PC-MDS programs to a single floppy diskette, thereby limiting the required system overhead. The minimal system requirements also result is increased computational speed. PC-MDS algorithms are fast, efficient and accurate, outperforming other programs on the market by as much as 50%. PC-MDS is used in Burke Institute's Advanced Multivariate Classes, as well as in the Professional Research Seminar Series, offered by the American Marketing Association. Again, thank you for your interest. If you have further questions, please don't hesitate to call. Scott M. Smith, Ph.D. Dept. of Marketing: 678 TNRB Brigham Young University PHONE: (801) 378-5569 Provo, Utah 84602 FAX: (801) 378-5984 EMAIL: smsmith@byu.edu Webpage: http://marketing.byu.edu PC-MDS MULTIDIMENSIONAL STATISTICS FOR THE PC PC-MDS 5.1 offers 22 multivariate statistical analysis programs for the IBM compatible personal computer, including Bell Laboratories multidimensional scaling, conjoint and cluster analysis programs and an SPSS (c) type interface for the BMD discriminant, regression and factor analysis programs. Version 5.1 introduces 3-D perceptual maps for MDS and correspondence analysis. All programs operate using easy to use pop-up menus. PC-MDS also supports a full screen editor for text and data manipulation. MULTIDIMENSIONAL SCALING: CONJOINT ANALYSIS PROGRAMS: MDPREF Preference Analysis MONANOVA Monotone ANOVA KYST Scaling Analysis TRADEOFF Attribute Tradeoff PREFMAP Preference Mapping CONJOINT Analyzer / Simulator INDSCAL Individual Differences PROFIT Property Fitting CLUSTER ANALYSIS: DATA MANIPULATION PROGRAMS: CLUSTER Howard Harris Cluster CASE5 Thurstone Case 5 HICLUST Hierarchical Cluster DISTRAN Distance Computation WARDS Ward's Method FMATCH Cliff Factor Matching EDIT Full Screen Editor CORRESPONDENCE ANALYSIS: AUTO. INTERACTION DETECTOR: CORAN French Algorithm AID Survey Research Ctr. CORRESP Carroll, Green, Schaffer MULTIVARIATE ANALYSIS: DISCRIM Discriminant Analysis REGRESS Regression Analysis FACTOR Factor Analysis FREQ Frequency Analysis Multidimensional scaling packages have in the past been difficult to use because of system incompatibilities, batch input and the mainframe environment. PC-MDS offers a friendly interactive format for even the most sophisticated analyses. PC-MDS is used in Burke Institute's Advanced Multivariate Classes, as well as in American Marketing Association Research Conference tutorials. For the research firm, PC-MDS means mainframe capability in your office. PC-MDS requires 640K of memory and is capable of handling problems larger than the original mainframe versions. Data may be analyzed and re-analyzed quickly and without additional costs of mainframe computer time. PC-MDS offers easy word processor access to the ASCII output text files. Plots and other output may be called into a word processor and moved into a report or manuscript by simply copying a block of text. Output may similarly be ranged and saved for spreadsheet and other graphics program input. For the educator, PC-MDS means ease of use and offers a total new dimension in student instruction. Pop-up menus and sample data sets create a sophisticated, yet user friendly package. Scaling, clustering, conjoint and AID analysis are discussed in the texts for most market research, modeling, and new product development classes. Now MBA and Undergraduate students can analyze and interpret these analyses. DOCUMENTATION: The PC-MDS package includes documentation describing the algorithms and input variables for the operation of each of the programs. Each program is accompanied by an example data set that is analyzed and explained in the documentation. (300+ pages) SYSTEM REQUIREMENTS: PC-MDS operates on any IBM PC or compatible having 640K of memory. Each program may be installed on a hard drive or diskette. PC-MDS is a DOS program with pop-up windows. It operates under DOS, and dos windows in Windows 3.1, 95, and Windows NT Version 4.0. Printer and VGA color graphics are recommended. PRICE: PC-MDS is sold on a single machine license basis. A one time site license purchase is also available. Site licenses include all machines at the license location. Single Copy: $ 495.00* Site license: $2000.00 * Plus Shipping and Handling: $5.00 U.S., $25.00 International There is no Shipping and Handling charge for a Site License TO UPDATE AN EARLIER VERSION: $ 200.00 Please make all payments in $US by check drawn on a US Bank. ORDER INFORMATION: Order by FAX, Telephone, or Mail Dr. Scott M. Smith Dept. of Marketing: 678 TNRB Brigham Young University PHONE: (801) 378-5569 Provo, Utah 84602 FAX: (801) 378-5984 Sample Programs at marketing.byu.edu email: smsmith@byu.edu Webpage: http://marketing.byu.edu PC-MDS PROGRAM DESCRIPTIONS AND SIZE LIMITATIONS MDPREF MultiDimensional PREFerence Scaling, performs an analysis of any type of dominance data for up to 100 stimuli (attributes) and 100 subjects (products). The program develops vector directions for preferences and the configuration of stimuli in a common space. (3-D Perceptual Maps) KYST Offers Kruskal Young Shepard Torgerson multidimensional scaling. KYST combines MDSCAL and TORSCA to include the powerful initial configuration of TORSCA along with the ability to rotate solutions to principal components. Proximity (similarity) data is used for up to 100 stimuli in 10 dimensions, with the limit of 5000 data points. KYST offers a variety of data input formats and distance measures. Monotone, polynomial and multivariate regression models may be derived. (3-D Perceptual Maps) PREFMAP Produces PREFerence MAPping analysis based on a generalization of the Coombsian unfolding model of preference. These models use a known configuration of stimuli to portray individual preference data. Up to 150 stimuli in 10 dimensions may be evaluated for 49 respondent groups. (3-D Maps) INDSCAL INdividual Differences SCALing produces an analysis of proximity (similarity) data. Program options include INDIFF, for scaling individual differences and CANDECOMP for performing the more general canonical decomposition of N-way tables. INDSCAL produces up to a 7 way solution for 10 dimensions. The product of the total number of levels must be less than 40,000. (3-D Perceptual Maps of both stimuli and subject group spaces). PROFIT PROperty FITting by Optimizing Nonlinear or Linear Correlation is a technique for fitting outside property vectors into stimulus spaces. Two types of property fitting: Max R, and non linear correlation, are performed. Performs separate linear or nonlinear, or both linear and non-linear regression for a set of 60 properties on 400 stimuli in 10 dimensions. (3-D Maps) MONANOVA MONotone ANalysis Of VAriance performs additive conjoint analysis of rank order data. This model handles full factorial designs and up to 12 factors can be handled with 100 levels per factor and a maximum of 4000 data points. TRADEOFF Performs a non-metric pairwise conjoint analysis. Utilities are produced for an attribute matrix of up to 10 levels by 10 levels. Up to 200 Attributes may be defined. Additive or multiplicative utilities may be specified. Utility importance values are derived. CONJOINT Performs a conjoint analysis of fractional factorial data. The program is compatible with design files from Bretton Clark's Conjoint Designer. This two stage program generates utilities, permits specification of part worth, vector, and ideal point models which are used in a simulator to generate choice share estimates using the Bradley-Terry-Luce, Logit and First Choice Models. 50 variables and 10,000 subjects. T-tests of differences between levels. CLUSTER Howard Harris CLUSTER analysis of a subject by variable matrix. K-means method minimizes within-group variance at each clustering level. 3000 subjects and 25 variables. HICLUST HIerarchical CLUSTering performs hierarchical cluster analysis on a 200 x 200 matrix of similarity or dissimilarity data (KYST or INDSCAL input data). Two methods are performed and cluster maps produced. WARDS Ward's method is a hierarchical agglomerative method of clustering. Ward's method optimizes the minimum variance within clusters using the within groups sum of squares. 300 subjects and 300 variables. Groupings and respondent's membership are produced. CASE5 Thurstone CASE 5 analysis computes scale scores from (1) raw paired comparisons data, (2) a frequency matrix of the number of times one stimulus was preferred to another stimulus, or (3) ranked data. Up to a 60 stimuli by 60 stimuli matrix. DISTRAN Computes a matrix of interpoint distances from a stimuli by variables data matrix. (100 stimuli and 100 attributes). FMATCH Cliff Factor MATCHing performs an orthogonal rotation of two matrices (such as MDS or factor analysis solutions) to achieve congruence. FMATCH will fit a 50 by 50 target matrix to any number of data matrices. AID Automatic Interaction Detector employs a non-symmetrical branching process based on ANOVA techniques (similar to a stepwise regression) to subdivide a sample into a series of demographic subgroups. Accommodates 60 variables for large data sets. CORAN CORrespondence ANalysis consists of techniques for mapping two way contingency tables for large data sets. Correspondence analysis is analogous to a principal components analysis of rows and columns of contingency tables (crosstabulation data). Correspondence analysis finds the best simultaneous representation of the categorical data. (Lebart, Morineau, Warwick algorithm) (3-D Perceptual Maps) CORRESP CORRESPondence analysis uses the Carroll, Green and Schaffer algorithm for mapping two way contingency tables for large data sets. This algorithm double centers the data producing more accurate representations of structure. Up to 100 rows and columns of contingency table (crosstabulation) data may be analyzed. (3-D Perceptual Maps) DISCRIM Interactive stepwise DISCRIMinant analysis. 50 variables and 32,000 subjects. Discrim was developed from the BMD statistical programs (Algorithms developed under the National Science Foundation). SPSS mainframe commands support extensive data manipulation. REGRESS Interactive stepwise multiple regression analysis. Up to 50 variables and 32,000 subjects. Regress was developed from the BMD statistical programs (Algorithms developed under the National Science Foundation). SPSS mainframe commands support extensive data manipulation capabilities. FACTOR Interactive FACTOR Analysis. Orthogonal and oblique rotation and output and interleaving of factor scores to the respondent data file are supported. Up to 50 variables may be analyzed for 32,000 subjects. FACTOR was developed from the BMD statistical programs (Algorithms developed under the National Science Foundation). SPSS mainframe commands support extensive data manipulation capabilities. FREQ Frequency analysis and descriptive statistics. Up to 250 variables may be analyzed for 32,000 subjects. Frequency was developed from the BMD statistical programs (Algorithms developed under National Science Foundation). SPSS mainframe commands support extensive data manipulation capabilities. EDIT A full screen ASCII editor with full search and replace, blocking and file management capabilities. Multiple files may be open simultaneously. * SPSS is a registered trademark of SPSS Inc.