Mendoza

Materials for Mendoza course, 29 June - 10 July

Monday

morning

  • outline (pdf, html)
  • overview (pdf, html, R, Rnw)
  • incantation for making Tinn-R copy and paste blocks correctly:
.trPaths <- paste(paste(Sys.getenv('APPDATA'),
      '\\Tinn-R\\tmp\\', sep=''),
        c('', 'search.txt', 'objects.txt','file.r',
       'selection.r','block.r','lines.r'), sep='')
  • readings: Platt 1964, Crome 1997, Gerber and Malhotra 2006 ("p-value" paper: Gerber, A., and N. Malhotra. “Can political science literatures be believed? A study of publication bias in the APSR and the AJPS.” Annual Meeting of the Midwest Political Science Association (2006): 12-15)

afternoon

Sys.setlocale("LC_TIME","English")

I think this should revert to normal (Spanish) when you start R the next time, but if not try
Sys.setlocale("LC_TIME","Spanish")

to switch back.

Tuesday

morning

afternoon

  • data entry and exploratory analysis (pdf, Rnw)

Wednesday

morning

afternoon

  • stochastic models (pdf)

projects

Thursday

Friday

topics.png

Monday

Tuesday

  • dynamic/state-space models (pdf)

Miscellaneous

  • Page 79 and following of Harrell, F. Regression Modeling Strategies. Springer, 2001 give some ideas for thinking about modeling with many predictors
  • Richards 2008, "Dealing with overdispersed count data in applied ecology" (J. Appl. Ecol. 218-227) gives a good description of QAIC etc.

Presentations

Miscellaneous

  • Page 79 and following of Harrell, F. Regression Modeling Strategies. Springer, 2001 give some ideas for thinking about modeling with many predictors
  • Richards 2008, "Dealing with overdispersed count data in applied ecology" (J. Appl. Ecol. 218-227) gives a good description of QAIC etc.
  • "the package formerly known as dRedging", by Kamil Bartón, is now MuMIn (Multi-model Inference). This package allows you to fit a model and all of its submodels (respecting marginality conditions, i.e. not fitting models where some interactions are present but their respective main effects or lower-order interactions are absent) and calculate AIC tables, do model averaging, etc.. This is a powerful, but potentially dangerous, tool: think before using! It was formerly available from his web page, now it is on R-forge; you can install it by saying
install.packages("MuMIn",repos="http://r-forge.r-project.org")

The web page says it is "pre-alpha" (i.e., experimental) but I think it should be fairly robust.

Spatial stuff

I never wrote down notes in any great detail: does anyone have some they'd like to volunteer?

References:

Unless otherwise stated, the content of this page is licensed under Creative Commons Attribution-ShareAlike 3.0 License