Featured

Have an animated 2014!

Animations can be a refreshing way to make a website more attractive, as we did exactly a year ago here.

Based on the Brownian motion simulated here, and using animation and ggplot2 R packages, we produced a fun welcome to the International year of Statistics, Statistics2013. The function saveMovie (please notice that there exists another alternative, saveGIF) allowed us to save it and finally publish it, as easy as that!

 
 
 
 
 
 
 
 
 
 
For this new year we have based our work on the demo(‘fireworks’) from the 2.0-0 version of the package animation (find  here a list of the changes in each version) and world from the package maps (speaking of maps, have a look at this great dynamic map by Bob Rudis!).

They also come very handy when trying to represent visually a dynamic process, the evolution of a time-series,etc.

As an example, in a previous post we  portrayed an optimisation model in which the values for the mean were asymptotically approaching the optimal solution, which was achieved after a few iterations. This was done with a for loop and again packages ggplot2 and animation.

There are many other examples of potential applications both in general Statistics – see this animated representation of the t distribution and these synchronised Markov chains and posterior distributions plots– and  in Biostatistics. Some examples of the latter are this genetic drift simulation by Bogumił Kamiński and this animated plots facility in Bio7, the integrated  environment for ecological modelling.

R package caTools also allows you to read and write images in gif format.

Finally, in LaTeX, \animategraphics from the animate package will do the trick; check this post by Rob J. Hyndman for further details.

It is certainly one of our new year’s resolutions to incorporate more animations in our posts, what are yours?

Advertisements
Featured

Interview with… Anabel Forte

foto_anabel

Anabel Forte Deltell, graduated in mathematics and statistics, holds a PhD in Statistics from the Univeristy of Valencia. Now she is a lecturer at the economics department of Universitat Jaume I at Castellón. Her research topics cover many fields of statistics as, for instance, spatial analysis or joint modeling of longitudinal (or panel) data and time-to-event data with her main interest beeing Bayesian model selection and, in particular, Objective Bayesian variable selection. Her dissertation (2011) entitled “Objective Bayes Criteria for Variable Selection” focuses on this last topic. She has published various articles in international journals and participated in many national and international meetings.

Email: forte@uji.es

 1.   Why do you like Biostatistics?

Researching in biostatistics is researching in real life and it makes me feel like I’m helping to improve people´s life conditions.

 2. Could you give us some insight in your current field of research?

Despite my main line of research being Bayesian model selection, I’m actually working in joint modeling of longitudinal data and time-to-event data. Nowadays many practitioners are moving to what it is called personalized medicine with medical decisions, practices, and/or products being tailored to the individual patient. In this scenario longitudinal data seems to be of great importance since it allows for the consideration of several sources of uncertainty including patient-specific variability. Moreover, quantifying the risk of suffering a certain event of a patient given its trajectory seems really sensible and can be done using joint modeling.

 3. Which are, in your opinion, the main advantages of being a researcher?

For me, being a researcher is simply a way of life. I can not imaging myself doing anything else.

 4. Your whole professional experience has been within the public sector and the University. How do you see the present and future of research in the Spanish public sector?

From my point of view if things do not change greatly, researching in Spain, at least in the public sector, has a really black future. As I see it, the only way researchers can manage to get money is to call attention of private investors. But private money can compromise the objectivity of research and at the same time avoid researching in some fields that are not so “attractive” for industry. Hence it is something that has to be studied carefully… but something has to be done, that´s for sure.

 5. What do you think of the situation of young biostatisticians in Spain?

As I see it, young biostatisticians have to make a great effort to show their enormous value for private companies as, for instance, food processing companies or health care related companies among others… And I think so because, pitifully, nowadays the Spanish public system can not assume all the biostatistical research that is needed.

 6. What would be the 3 main characteristics or skills you would use to describe a good biostatistician?

For me it is really important to have a good mathematical/probabilistic base and to be well organized when working. But the most important of all is enjoying research.

 7. Which do you think are the main qualities of a good mentor?

For me, a good mentor is someone that supervises your work and is always there for you but at the same time he or she should give you some space for you to fail and learn… in other words someone that teaches you what researching really is.

Selected publications:

  • Francisco Pozo-Rodríguez; Jose Luis López-Campos; Carlos J. Álvarez-Martínez; Ady Castro-Acosta; Ramón Agüero; Javier Hueto; Jesús Hernández-Hernández; Manuel Barrón; Victor Abraira; Anabel Forte; Juan Miguel Sanchez Nieto; Encarnación Lopez-Gabaldón; Borja G. Cosío; Alvar Agusti. Clinical Audit of COPD patients requiring hospital admissions in Spain: AUDIPOC Study.Plos One. 7 – 7, (USA): 2012.
  • M. J. Bayarri; J. O. Berger; A. Forte; G. García-Donato. Criteria fo Bayesian model choice with application to variable selection. Annals of Statistics. 40 – 3, pp. 1550 – 1577. (USA): 2012.
  • C. Armero; A. Forte; A. López-Quílez. Geographical variation in pharmacological prescription. Mathematical and Computer Modelling. 50, pp. 921 – 928. (UK): 2009.
  • Allepuz; A. López-Quílez; A. Forte; G. Fernández; J. Casal. Spatial Analysis of Bovine Spongiform Encephalopathy in Galicia, Spain (2000-05). Preventive Veterinary Medicine. 79 – 2, pp. 174 – 185. (Holland): 2007.
  • López-Quílez; C. Armero; A. Forte. Geographical variation of Pharmacological prescription with Bayesian Hierarchical models.Value In Health. 10 – 6, (USA): 2007.