| |
:: Course SummaryThis 2-day workshop is a logical continuation of the 3-day session on Multivariate Data Analysis 1 : Classical Methods. Through cases studies selected from different fields of activity (R&D, marketing, business) participants will discover original applications of classical multivariate methods, judicious combinations of methods as well as advanced methods derived from classical ones to address complex research questions.
Participants who have attended the previous session will be given the opportunity to refresh their knowledge and to deepen their understanding of multivariate methods by learning new approaches and strategies.:: Learning ObjectivesUpon completion of this workshop, participants will be able to:
- Harness the power of advanced multivariate data analysis methods;
- Determine which multivariate methods or combination of methods can be used for a given study objective;
- Carry out the analysis using their own software;
- Extract the pertinent or relevant information from the output provided by the software;
- Interpret the numerical and graphical results.
- Summarize and communicate the information obtained
- Produce insightful graphical summaries
:: Target AudienceThis workshop is intended for non-statisticians who possess a good knowledge of classical multivariate techniques interested in learning how to efficiently combine the, to extract as much information as possible in their data. Is is also intended for statisticians looking for an applied workshop and who wish to learn more about the potential of multivariate techniques and efficient ways of combining them.:: PrerequisiteParticipants must possess a good working knowledge of Principal Component Analysis (PCA), Factor Analysis, Correspondence Analysis, and Cluster Analysis or they must have attended one of the following the training sessions :
:: Notes and Other InformationSupported software are: R, S-Plus, SAS, SPSS, XLSTAT. If you use another software and you wish to attend this workshop, contact us.
Course Specifications for Registered Participants :
- We invite you all to bring your laptop containing your multivariate analysis software of choice and, optionally, your own data as well. This will help you get the most out of the workshop by seeing the immediate application of statistical techniques using your own data and software.
- Please let us know what software package you are planning to use so that we may plan accordingly.
- We will provide sample datasets for those you who prefer not to bring your own data.
The course venue for the June 2012 edition is Hotel des Coutelliers, a boutique hotel among the top-rated on TripAdvisor, located in the heart of the old city. Note that a few rooms have been reserved for Creascience and are available on a first come-first serve basis at a discounted rate of 110CAD per night (sgl occ. exc. app. tx). Breakfast and a wireless internet connection are included in this price. Alternatively, there are many other accommodation options within walking distance of the course venue.
Registration Deadline: May 15, 2012 | | |
:: Course ContentThis workshop, logical continuation of the 3-day session on Multivariate Data Analysis 1 - Classical Methods, covers methods derived from classical multivariate methods and gives participants the opportunity to discover original applications with the help of judicious combinations.
Here are some of the original applications that will be covered during the workshop:
- Principle of discriminant analysis to differentiate between groups of objects;
- Multivariate outlier detection;
- Selection of a subset of representative objects from a population;
- Exploration and visualization of the underlying product preference structure of consumers;
- Use of canonical correlation analysis to establish the most important relationships between two sets of data;
- Introduction to the analysis of curves (spectral data and other complex responses)
- Quick homogenization of a set of variables to build groups;
- Prediction to which class new objects belong;
| |
