Preference Modelling using Partial Least Squares in XLSTAT, July 16th, 2020
This short course delivered online will show consumer scientists how to use partial least squares in XLSTAT for relating consumer acceptability to sensory/analytic measures.
Preference Modelling using Partial Least Squares in XLSTAT
The remote learning package consists of a live interactive lecture delivered online by one of the Qi Statistics team, together with workshops for you to complete in your own time (solutions and data sets provided), plus a pdf of the lecture notes and email support whilst you are completing the workshop after the course for up to 2 weeks. More than just a recorded webinar and as close to face-to-face training as it can be!
This module follows on from the Preference Mapping module, (held on 25th June) and focuses on using more advanced methods based around regression modelling in XLSTAT for relating consumer acceptability to sensory/analytic measures.
The course content is as follows:
- Partial Least Squares (PLS) explained
- Modelling Liking v sensory profile data or analytic data
- PLS approach compared to PCA based PREFMAP
- PLS measures explained (VIP, Q Squared), are PLS confidence intervals reliable?
- RV coefficients as a investigative tool
- Using PLS to link sensory and analytic/instrumental data
- An extensive set of exercises with solutions (explanation of statistics and interpretation of maps) will be included as part of the training.
To be able to perform the exercises you will need a full demo-version of XLSTAT (or licensed copy of XLSTAT Basic + Sensory Methods module).
Length of Session:
Approx 1hr + workshops (also available at alternative time for other time zones – see course listings)
A basic understanding of statistics is assumed i.e. knowledge of statistical terminology and summary measures such as mean, variance, standard deviations. It would also be expected for the participant to have a basic knowledge of preference mapping techniques or have attended the previous webinar on preference mapping (recording may be available to purchase online).
Dr. Hal MacFie is a statistician by training, with an international reputation in the areas of product assessment and consumer research. He is a Visiting Professor at the Universities of Reading and Nottingham and was for 20 years a senior Editor of Food Quality and Preference.