Stats beerz - An ad hoc aggregation of statistical and resources
A humble tribute to William Sealy Gosset (a.k.a. Student) who created the t-test to improve the quality of Guinness. I always knew statistics was good for something.
This is an informal get-together of SFU ecologists and statisticians to bounce around ideas on graphing, coding and statistics; from ANOVAs to Z-scores.
Here are some of our favorite things...
This is an informal get-together of SFU ecologists and statisticians to bounce around ideas on graphing, coding and statistics; from ANOVAs to Z-scores.
Here are some of our favorite things...
New R stuff
Why I want to write nice R code, Daniel Falster
Tutorial for dplyr by Hadley Wickham Learn R & Become a Data Analyst by DataCamp R Library: Contrast Coding Systems for categorical variables Analysis of community ecology data in R program, workshop and excercises R training package SWIRL designed to teach both R and statistics simultaneously and interactively and blog post 91 two minute tutorials on how to programme in R from www.twotorials.com R Markdown in RStudio: http://www.rstudio.com/ide/docs/authoring/using_markdown Sean Anderson's notes on debugging R functions: http://seananderson.ca/2013/08/23/debugging-r.html The graphical debugger just announced for RStudio: http://www.rstudio.com/ide/docs/debugging/overview Programming
Getting started on github by Roger Dudler
Interactive notebooks: Sharing the code, by Helen Shen - I-Python notebook demo Progress on the ggviz cookbook Interactive plotting with Manipulate in R If you watch only one Youtube video on programming watch Greg Wilson on SciPy Back to basics
Neat ways to look at correlations in corrplot.
Why ANOVA and regression are the same Analysis of variance--why it is more important than ever. Andrew Gelman Scientific method: Statistical errors (P values, the 'gold standard' of statistical validity, are not as reliable as many scientists assume) Regina Nuzzo Statistics for biologists – A free Nature Collection If you really have to use P-values then read: Johnson VE. 2013. Revised standards for statistical evidence. Proceedings of the National Academy of Sciences 110: 19313-19317. Paul Allison of Statistical Horizons When multicollinearity may not be a problem and VIFs Dates in R - sequences of dates from Inside-R Graph prettification & colour
A Compendium of Clean Graphs in R
Tools for exploring distance matrices using circular plots. http://martinwestgate.com/code/circleplot A ggplot cheatsheet by Sharon Machlis Bad grafs by Daniel Pauly We often use animal and plant silhouettes to prettify our graphs. Here are some useful OA sources Integration and Application Network, University of Maryland Center for Environmental Science Phylopic How to improve pie charts by www.darkhorseanalytics.com Colour explorers Adobe Kuler iWantHue Colourlovers Subtleties of Color (Part 1 of 6) Robert Simmon, Nasa Dear NASA: No More Rainbow Color Scales, Please, Drew Skau Rainbow Color Map (Still) Considered Harmful, David Boreland and Russel M. Taylor II The amazzzing ggplot2 book by Hadley Wickham Extracting data from graphs
How to digitise graphs in R
Twitter content analysis in R, cool slideshare by Yangchang Zhao Getting data from an image using R Data thief III, Java app for Windows, Mac, Unix GraphClick for mac, $7.99 from Apple store Plot Digitizer for Windows, free Model fittingDRAFT r-sig-mixed-models FAQ
On Some Alternatives to Regression Models by Arthur Charpentier @freakonometrics Plotting model coefficients with SJplot package spj.lm centering predictors when model averaging across interactions from Sean Anderson http://seananderson.ca/2014/07/30/centering-interactions.html interpreting coefficients for log-transformed response data from Sean Anderson Rolling your own hurdle model to model continuous-positive data with zeros from Sean Anderson Schielzeth H. 2010. Simple means to improve the interpretability of regression coefficients. Methods in Ecology and Evolution 1: 103-113. Markov Chain Monte Carlo & Bayesian nastiness
Zero-inflated GLMMS | check out the "Owl nestling negotiation" section of the non-linear modeling projects at NCEAS - https://groups.nceas.ucsb.edu/non-linear-modeling/projects
Andy Cooper's illustration of MCMC for linear regression: http://www.youtube.com/watch?v=7GMxM-zLoHo R code and 3-dimensional illustration of MCMC: http://mbjoseph.github.io/blog/2013/09/08/metropolis/ MCMC diagnostics in ggplot: http://www.r-bloggers.com/ggmcmc-diagnostic-plots-for-mcmc-with-ggplot2/ Multivariate statistics
How does k-means clustering work? Neat visualisation by Aleksey Nozdryn-Plotnicki
Legendre, P. 2005. Species associations: the Kendall coefficient of concordance revisited. Journal of Agricultural, Biological, and Environmental Statistics 10:226-245. Principal component analysis (PCA) with options for scalings and output by Pierre Legendre's R code Redundancy Analysis (Pascale Gibeau) Pierre Legendre's website is: http://adn.biol.umontreal.ca/~numericalecology/indexEn.html R code for RDA and do the figures too, and they are really neat. They test the relationships with permutations, which is much easier with ecological data (no need to have normal data, and less worries about spatial and temporal autocorrelation). Have a look (his website is bilingual, and his codes are in English). He also has a great book (Numerical Ecology) that goes over many multivariate techniques, including ordinations (PCAs, etc.) and canonical ordinations (RDA, CCA). |
If you'd like to attend statz-beerz meetings sign up to get announcements through the stats-beerz mailing list: http://maillist.sfu.ca/ (You'll need to be registered with your @sfu.ca address.)
