Publications by Wingfeet

JAGS model Fe concentration in rainwater including values below detection level

01.12.2013

In my previous post I ignored the fact that some data are below the detection level. I would not know how to handle those in a mixed model from lme4 or nlme. However, JAGS can handle these values. Next to that I kept the usual independent variables, such as random effects for location, machine etc.DataData has been acquired and proces...

9924 sym 2 img

Fukushima region radiation decrease

08.12.2013

Fukushima is still a topic which gets headlines, and somewhere in a comment was a link to actual and historical radiation data: Fukushima prefecture radioactivity measurement map.  It takes a bit of time to load but then you have a map of the Fukushima region with current radiation levels at measurement stations. Somewhere on that w...

4474 sym 4 img

plotting y and log(y) in one figure

15.12.2013

Sometimes I have the desire to plot both on the linear and on the log scale. To save space just two figures is not my solution. I want to reuse the x-axis, legend, title. This post examines possibilities to do so with standard plot tools, lattice and ggplot2.DataData is completely artificial.library(ggplot2)library(lattice)datastart �...

2962 sym 6 img

Sinterklaas and Santa Claus gave presents

27.12.2013

In December my nice little netbook acquired a shiny new openSUSE 13.1, including all the goodies I might want to use in 2014; R, Stan, Julia and Jags. So here is what you might get when you set up a computer for statistics today.JagsI have been using Jags for quite some time because I had the (incorrect) impression openBugs developmen...

4450 sym 2 img

Le Monde puzzle 847 in Julia

04.01.2014

This week I wanted to play around with Julia and exporting the results. I found http://xianblog.wordpress.com/2013/12/29/le-monde-puzzle-847/ to be just the right size to play around with.CodeA function to check if a triplet has the desired property In [1]: function lemonde847(xx) a=[xx[1],2] b=[xx[2],8] c=[xx[3],4] sum(kron(c,kron(...

2120 sym R (860 sym/9 pcs)

Converting a JAGS model to STAN

11.01.2014

For my first experience with STAN I wanted to convert my last JAGS program into STAN. This was a bit more difficult than I expected. The JAGS program was Fe concentration in rainwater including values below detection level.DataData has been explained before. The only difference is that by now I am reading the data using read.xls from ...

11934 sym

Convering plots to data

18.01.2014

It is a problem which occurs ever so often in applied work, you have a plot, but you want the data. There are at least two programs which can help you there; PlotDigitizer and Engauge Digitizer. I got both on my openSuse machine. Both are available for Windows, for Mac there are only older versions of Engauge.I tried these programs on...

2685 sym 2 img

Converting plots to data II

26.01.2014

There were quite some reactions on lasts week’s post converting plots to data. Two additional programs were mentioned; WebPlotDigitzier and DataThief. Note that google or wikipedia has a bunch of alternatives. Datathief is shareware, a java program, so can run on Mac, Windows and Linux. WebPlotDigitizer is, as one might guess, web b...

2545 sym 2 img

Bayesian analysis of sensory profiling data

02.02.2014

I looked at Bayesian analysis of sensory profiling data in May and June 2012. I do remember not being totally happy with the result and computations taking a bit more time than I wanted. But now it is 2014, I can use STAN and I have been thinking about the model I want a bit more. Hence a fresh start.DataData is the chocolate data fro...

5754 sym 4 img

Bayesian analysis of sensory profiling data, part 2

09.02.2014

Last week I made the core of a Bayesian model for sensory profiling data. This week the extras need to be added. That is, there are a bunch of extra interactions and the error is dependent on panelists and descriptors.Note that where last week I pointed to influence of Procrustes and STATIS in these models, I probably should have mentioned Per Br...

6711 sym 8 img