Category Archives: Uncategorized

ggmcmc – diagnostic plots for MCMC with ggplot2

Daniel Hocking:

Plotting MCMC output with ggplot using ggmcmc

Originally posted on JAGS News:

Xavier Fernández i Marín, who maintains the jags package on Gentoo Linux, writes to tell me he is developing the R package ggmcmc . This package is for visualizing Markov Chain Monte Carlo output using ggplot2 graphics and  should complement the existing plots for base and lattice graphics provided by coda . A comparison of all three graphical styles is given below.

View original 83 more words

Postdoctoral Research Position: Extreme event effects on stream fish and salamanders

There is a new postdoc position available with the lab group I work in with Ben Letcher and Evan Grant. Below is a copy of the posting. I’ve just been part of the group for a couple months but I love it. The people are social and great to work with. It’s a high energy group working on lots of really neat projects, but it doesn’t have the high pressure, negative atmosphere that develops in some big academic labs. The group certainly won’t be featured in PhD Comics. Feel free to contact me if you have questions about the group and lab dynamics (project-specific questions will have to be directed to Ben Letcher).

http://wfscjobs.tamu.edu/jobs/extreme-environmental-effects-on-demography-of-stream-fish-and-salamanders-massachusetts/

Climate change will likely increase the frequency and severity of extreme events. Understanding and forecasting biological response to extreme events represents a key challenge to applied ecology. This project involves incorporating forecasts of extreme events into population models for stream fish and salamanders in the northeastern US. Existing models include detailed demographic models (integral projection models), abundance models and occupancy models. Extreme events in streams mainly result in floods and droughts, so models will account for effects of these drivers on vital rates and probabilities of local extinction. Models will primarily rely on existing data, but there is an opportunity to conduct flume tests of the effects of environmental drivers on vital rates if necessary. The post-doc will collaborate directly with a hydrologic modeler, and indirectly with a large team of environmental and biological modelers.

The USGS Conte Anadromous Fish Research Center in Turners Falls, MA seeks a postdoctoral research associate to develop demographic models for species response to extreme events. The project is part of a larger US Department of Interior effort to understand effects of hurricanes on the northeastern US. Preferred start date is September 1, 2014.

Qualifications
Competitive candidates will have skills in demographic modeling (projection matrices, abundance and occupancy models), estimation (particularly in a Bayesian context), and optimization, with previous experience working on collaborative research projects and familiarity with headwater streams. We are especially interested in applicants with experience linking environmental drivers to demographic models. The successful candidate will also have excellent writing and personal communication skills. Applicants are expected to have earned a Ph.D. degree in a relevant discipline, preferably within the last 2 years.

Salary is $52K per year plus a health care allowance, with support available for 2 years. Travel funds will be provided to attend regional workshops during model development, and to attend professional conferences.

Send a letter describing your background and experiences as they relate to this position, and a CV with names and contact information for three references to Ben Letcher (bletcher@usgs.gov). To receive full consideration, send application materials by 31 March 2014.

2013 in review

Now is the time of year where we all do a little navel gazing, so here is my look back at 2013. I’ll start with a bit of summary about the blog. I started this blog with a few purposes 1) to increase my broader impacts for grants, 2) provide resources for others because I’ve found some much helpful R code on other people’s blogs, 3) foster interactions with researchers beyond my university (where I sometimes feel a bit isolated), and as a bonus I’ve found that writing many of these posts helps me clarify my own thinking on various topics.

The WordPress.com stats helper monkeys prepared a 2013 annual report for this blog.

Here’s an excerpt:

The concert hall at the Sydney Opera House holds 2,700 people. This blog was viewed about 15,000 times in 2013. If it were a concert at Sydney Opera House, it would take about 6 sold-out performances for that many people to see it.

Click here to see the complete report. Now if only most of those weren’t bots trying to sell generic viagra and mine my data I’d be all famous. At least some of the visits were from actual people interested in silly things like statistics, R, and amphibian ecology. Apparently, my Valentine’s Day post, No Statistical Panacea, Hierarchical or Otherwise, was sufficiently interesting to delay 214 trips to Hallmark. It even garnered positive responses from two of my academic heros, Ben Bolker and Brian McGill. Happy Valentine’s Day to me!

My 5 most popular posts were

1 High Resolution Figures in R
2 GEE QIC update
3 Stan for Bayesian Analysis
4 No Statistical Panacea, Hierarchical or Otherwise
5 Knitting beautiful documents in RStudio

I really should redo the post on producing high resolution figures in R. I wrote it as I learned how to do it and it shows my exploration with trials and errors. I should make a new post with better information and link to it at the top of the old post to be more helpful to people since it is getting a fair amount of traffic. I guess that’s something for an early 2014 post.

I also published 3 papers this year, have a 4th in press, and another 2 in review so it was a successful year from that standpoint.

Hocking, D. J. and K. J. Babbitt. In review. Amphibian Contributions to Ecosystem Services. Herpetological Conservation and Biology. (Open Access)

Anderson, T. L., D. J. Hocking, C. A. Conner, J. E. Earl, E. B. Harper, M. S. Osbourn, W. E. Peterman, T. A. G. Rittenhouse, and R. D. Semlitsch. In review. The influence of priority effects on metamorph traits and recruitment of two pond-breeding salamanders. Oecologia.

Hocking, D. J. and K. J. Babbitt. Accepted. The role of red-backed salamanders on ecosystem functions. PLOS One.

Hocking, D. J. 2013. Comparing the influence of ecology journals using citation metrics: making sense of a multitude of metrics. Ideas in Ecology & Evolution. 6: 55-65 . DOI: 10.4033/iee.2013.6.17.f  (Open Access)

Hocking, D. J., S. A. Callaghan, K. J. Babbitt, and M. Yamasaki. 2013. Comparison of silvicultural and natural disturbance effects on terrestrial salamanders in northern hardwood forests. Biological Conservation. 167: 194-202. DOI: 10.1016/j.biocon.2013.08.006

Hocking, D. J., G. M. Connette, C. A. Conner, B. R. Scheffers, S. E. Pittman, W. E. Peterman, and R. D. Semlitsch. 2013. Effects of experimental forest management on a terrestrial, woodland salamander in Missouri. Forest Ecology and Management 287:32-39. http://dx.doi.org/10.1016/j.foreco.2012.09.013. Please email me if you would like a copy. You can read my brief research summary here.

I also started using git and GitHub for version control and backup of my projects. I’m still learning how to use branches and such, but it has been a useful tool for my research and my teaching. I also started using R markdown and knitr, which has been great for teaching. I also started using Twitter (@djhocking) and have had some great interactions with researchers from around the world and found links to great resources that I otherwise might not have encountered.

The end of 2013 also marks the end of my time at the University of New Hampshire. After 5 years of grad school and 1.5 years as a postdoc, I will be starting a new postdoc position at UMass-Amherst in January. I am very excited about the new position and associated project. I will be working with a great group of people at the Northeast Climate Science Center, Forest Service, and USGS to model climate change effects on brook trout and stream salamander populations.

AnnalisseFinally and most importantly, 2013 was a tremendously exciting year owing to the birth of my daughter, Annalisse. Watching her grow and develop over the last 8 months has been the most amazing experience of my life. Despite running taking a backseat to family and work this year, I did run my first marathon finishing in 2:25:49 (5:33/mile) for 10th place. It was a great experience with my wife and our friend pushing strollers to half a dozen places on the course to cheer for me and his wife.

All and all 2013 was pretty good but I am looking forward to new challenges and adventures in 2014!

In Praise of Exploratory Statistics

If you haven’t seen it, you should definitely check out the latest post by Brian McGill at Dynamic Ecology. It’s a great post on the use of exporatory statistics. While it may be impossible to get an NSF grant with an analysis framed in terms of exploratory statistics, reviewers should definitely be more open to their appropriate use. Is is wrong to force authors to frame exploratory analyses in a hypothetico-deductive framework after the fact. Brian suggests that there are three forms of scientific analysis:

  1. Hypothesis testing
  2. Prediction
  3. Exploratory

and only one of these can be used for a single data set. My favorite quote is,

“If exploratory statistics weren’t treated like the crazy uncle nobody wants to talk about and everybody is embarrassed to admit being related to, science would be much better off.”

There’s lots of good stuff in his post, so I recommend spending 10 minutes of your day to read through it.

JSON, GeoJSON, Bash, and R

Follow

Get every new post delivered to your Inbox.

Join 40 other followers