Publications by Wingfeet
Tuning GAMBoost
This post describes some of the simulation results which I obtained with the GAMBoost package. The aim of these simulations is to get a feel what I should tune and what I should not tune with GAMBoost. SetupIn the GAMBoost package one can tune quite a number of parameters. I have looked at tuning bdeg (degree of the B-spline basis to be used fo...
4356 sym 2 img
Bayesian ANOVA for sensory panel profiling data
In this post it is examined if it is possible to use Bayesian methods and specifically JAGS to analyze sensory profiling data. The aim is not to obtain different results, but rather to confirm that the results are fairly similar. The data used is the chocolate data from SensoMineR and the script is adapted from various online sources ...
7505 sym R (7166 sym/4 pcs) 4 img
Multiplicative effects in sensory panel data
In a previous post I used JAGS to build the Bayesian equivalent of a two-way ANOVA. Effects were determined of products, panelists and their interaction. In this post this model will be rebuild to provide a more simplified and advanced model. The interaction between panelists and products is removed, which is the simplification. A mul...
7971 sym 4 img
Extending the sensory profiling data model
In this post I extend the multiplicative Bayesian sensory profiling model with effects for rounds and sessions. Is is not a difficult extension, but it brings the need for informative priors into the model. I do believe round and session effects exist, but, they are small. The Bayesian paradigm allows to employ small directly in the m...
13761 sym 2 img
A complete Bayesian model for sensory profiling data
In this post I will try to add an important parts in the sensory profiling model I have been building. This concerns the question: ‘Are all panelists equally reproducible?’. Obviously the answer is no, some are better than others. From this observation stems the approach in which under performing panelists are removed prior to the...
8937 sym 8 img
Simulation in the profiling model
In this post I try to make a small simulation of the sensory (flavour) profiling data, and examine if the parameters of simulated data can be retrieved by the Bayesian model build in the previous posts.The conclusion is that it is difficult, the amount of uncertainty is too large for parts of the data. Especially, the multiplicative e...
4710 sym 10 img 1 tbl
Creating Williams designs with even number of products
A Williams design is a special Latin square with the additional property of first order carry over (each product is followed equally often by each other product). In R the package crossdes can be used to create them. > williams(4) [,1] [,2] [,3] [,4][1,] 1 2 4 3[2,] 2 3 1 4[3,] 3 ...
3622 sym 2 img
Williams designs with 5 products
In a previous post I created small Williams designs for an even number of products. This worked very well, also because the number of permutations could be restricted significantly due to symmetry. Unfortunately this does not work so well with an odd number of products. The symmetry is not so obvious. In fact, using brute force I can ...
11270 sym 6 img
Trying Julia
In my previous post I tried building Williams designs in R. Since that code was running a bit slow, this was an ideal test for Julia. Big enough to be at least slightly realistic, small enough that it is doable.I am very impressed. Almost twenty fold speed increase, even though this was the best I could do in R, the most naive way pos...
2976 sym 2 img
Random and fixed effects in sensory profiling
I am reading Introduction into mixed modelling by N.W. Galway. It is partly a repeat of things I know, but I expect to use mixed models quite a lot the coming time, so it is good to repeat these things.My problem with this book is a sensory example in chapter 2. It is profiling data, but with some twists, as happens in real life. He ...
4137 sym 2 img 2 tbl