Tuesday, January 13, 2015

qualitative or mixed data analysis

LinkedIn даёт пакет для такого анализа.
#Rstats #research My brief survey of R packages for qualitative (or mixed) data analysis. http://qr.ae/6auZw

Javier, I appreciate the suggestion and the link. In my post, I've tried to limit the discussion to R packages that are much more focused on qualitative data analysis than many others. I understand that "lme4" and many other packages contain functionality that deals with qualitative (in a form of categorical) or mixed data, but I was trying to limit the scope of the material for better focus. More importantly, mixed-effects modeling is IMHO inherently quantitative approach. Nevertheless, this is an interesting topic, which I plan to become acquainted more with in the future. Thank you again and feel free to connect!

Check also this package: http://factominer.free.fr/advanced-methods/multiple-factor-analysis.html

Ahmed, I'm aware of "factominer" and some other FA packages. However, I tend to think of them mostly as packages for quantitative data analysis, despite their support for categorical data. Please see my points in the comment for Javier above. Anyway, I appreciate your comment.

Thursday, November 20, 2014

Forecasting mortality, fertility, migration and population data

The demography package for R contains functions for various demographic analyses. It provides facilities for demographic statistics, modelling and forecasting. In particular, it implements lifetable calculations; Lee-Carter modelling and variants; functional data analysis of mortality rates, fertility rates, net migration numbers; and stochastic population forecasting.

and more

Sunday, April 7, 2013

Top 10 tips to get started with R

(This article was first published on mages' blog, and kindly contributed to R-bloggers)

  1. Be motivated. R has a steep learning curve. Find a problem you can't solve otherwise. E.g. plotting multivariate data, a statistical analysis for which an R function exists already.
  2. Download and install R. Get to know the R console. Learn how to install additional packages, how to access the history, how to use auto completion and open the help system. Review the R Installation and Administration manual and check out the free books section on CRAN. 
  3. Get familiar with the R help files. They can appear cryptic at the start, but there is a structure to them. Read and re-read a couple of help files again and again. Look out for the input and output sections, execute the examples, run the demos, e.g.demo(graphics). Subscribe to R-help and read questions and answers, check outstackoverflow, follow blogs. Search with Rseek.org
  4. Learn how to get your data into R. The easiest way is usually via a CSV-file (CSV=comma separated values), using read.csv. Look into XLConnect, if you have to deal with spreadsheet files. Move on to write queries against data bases, e.g. using RODBC. Skim through the R Data Import/Export manual. 
  5. Try to understand the different data types in R and how to modify them. What are the differences between a matrix and a data frame? What is a factor? What is a list? Think about the different use cases. Review the Introduction to R manual.
  6. Do charts! Lots of charts. They are rewarding and keep you motivated. Be inspired by the R Graph Gallery. Check out the following packages: latticeplotrixggplot2,deducergoogleVis.
  7. Learn how you can modify and reshape data in R and apply functions on subsets using by, apply, lapply, avereshapesweep, with, within, etc. Set aside a weekend to think about these functions. 
  8. Write your R code into files instead of typing it all into the R console. Use anintegrated development environment (IDE), e.g. ESS EmacsRStudioStatET Eclipse.
  9. Understand the concept of functions. Write a function, which gives "Hello World" back. Modify it, so it has an input argument NAME and it prints "Hello NAME". Review the code of existing R functions. Copy from existing code.
  10. Document your code! Start your code by explaining what you want to achieve and only code that much, then write down the next step in plain English and code again. How will you know that your code does what you want it to do? Testing can help. Think your about your code style and how you will be versioning your files.

Bonus tip

To leave a comment for the author, please follow the link and comment on his blog:mages' blog.
R-bloggers.com offers daily e-mail updates about R news and tutorials on topics such as: visualization (ggplot2Boxplotsmapsanimation), programming (RStudioSweaveLaTeX,SQLEclipsegithadoopWeb Scraping) statistics (regressionPCAtime series,ecdf,trading) and more...

Tuesday, February 26, 2013

Statistics for the Social Sciences

страничка jfox at mcmaster.ca на кране
с короткими рекомендациями
любопытно, что этому блогу уже больше 2х лет, начал 3 февраля 2011

Monday, February 25, 2013

couple of resources

Rob Kabacoff's Quick-R accessing the power of R, and
RDataMining.com: R and Data Mining

still does not work

Это про Rcmdr
Удалось установить только RStudio, a кроме упомянутого Rcmdra ещё хочу Rattle, который тоже не встаёт :(

вопросы задавал в ЖЖ и
в LinkedIn
ответы есть, но, тем не менее, воз и ныне там, а нужда уже подпирает

устанавливал по оригинальной инструкции, есличо

Thursday, December 13, 2012

meaning forecast

robjhyndman.com пишет про новый R пакет прогнозирования как бе, но пишет следующим образом:
The accuracy func­tion now works with both time series fore­casts and cross-​​sectional forecasts. то-есть, по-руски прогнозирование -- это предвидение по времени, а попе н достансге -- совсем не обязательно и вводит нас (если на демографическом образовании строицо) в мир indirect techniques

Wednesday, December 5, 2012

Multivariate regression analysis using categorical explanatory variables

LinkedIN начал новую нитку с этой назвой

Which package is useful to handle categorical variables in a multivariate regression analyisis and where may I find the tutorials? Every information that I find is about quantitative data.
I am currently using a lm function to find the best predictor for a quantitative response variable but I am afraid, it is not the best approach. I also do not want to use dummies.

Thursday, October 11, 2012

research vs grants

Rob J Hyn­d­man :
I don’t think telling peo­ple what research areas are impor­tant is very help­ful. The best research comes about when peo­ple are curi­ous and pas­sion­ate, and take a sur­pris­ingly dif­fer­ent per­spec­tive from those who have gone before.
from here
действительно, как раз сейчас пытаюсь ответить на вопросы какбе грантодателей, 1 из х:
Существует ли в стране политика, программа и/или стратегия, направленная на решение вопросов сексуального и репродуктивного здоровья и репродуктивных прав, которая(ые) находится в процессе разработки или исполнения?
ответить надо да/нет

Thursday, April 26, 2012

Tutorials from Universities Around the World

Online tutorials for R programming, Statistics and Graphic

Here is a list of FREE R tutorials hosted in official website of universities around the world. The tutorials are listed  in no particular order, actually based on when I have discovered it. They will be categorised soon. Please kindly suggest me other university-hosted online R tutorials by email to me@pairach.com.
  1. University of California at Davis,
    Getting Started with the R Data Analysis Package
    by Professor Norm Matloff
  2. Clarkson University,
    R Tutorial
     Kelly Black
  3. York University,
    Getting started with R
  4. University of Waterloo
    R Tutorial For A WINDOWS Environment (WindowsUnix)
  5. University of California at Los Angles, UCLA,
    Resources to help you learn and use R
  6. University of California at Riverside.
    Programming in R
  7. University of Illinois,
    A Brief Introduction to R
  8. University of Texas at Austin,
    R Tutorial Videos
    by Brandon K. Vaughn
  9. University of California at Berkeley,
    An Introduction to R (PDF)
    by Phil Spector
  10. University of California at Santa Babara,
    R Programming Resource Centre
    by National Center for Ecological Analysis and Synthesis
  11. Chiang Mai University,
    Econometrics with R (in Thai)
    by Pairach Piboonrungroj.
  12. University of Carnegie Mellon,
    A Tutorial: Some Fundamentals of R.
    by Bruce E. Trumbo
  13. University of Illinois State,
    R Tutorial.
    by Dong-Yun Kim
  14. University of MacMaster,
    Introduction to the R Statistical Computing Environment.
    by John Fox.
  15. University of Princeton,
    Introducing R.
    by Germán Rodríguez
  16. University of Amsterdam,
    How to draw graphs with R,  [Graphics]
    by A.M. (Angelos-Miltiadis) Krypotos
  17. University of North Texas,
    Do it yourself – Introduction to R [Intro]
  18. University of Warwick,
    R programming page [Biosciences: Molecular Organisation and Assembly in Cells]
    by Peter Cock.
  19. University of Illinois at Urbana-Champaign
    R tutorial for Applied Econometrics
    by Prof. Roger Koenker
  20. Coastal Carolina University
    R tutorials [General]
    by  William B. King
  21. University of Colorado Denver
    R Tutorial [General]
    by  Stephanie Santorico and Mark Shin
  22. Stanford School of Medicine (Biomedical Informatics)
    R Tutorial [VDO on Introduction + Translational Bioinformatics]
  23. Harding University
    Producing Simple Graphs with R [basic graphic e.g., line, bar, hist, line]
    by Frank McCown
  24. University of Kentucky, Department of Statistics
    Use Software R to do Survival Analysis and Simulation [pdf]
    by Mai Zhou

Sunday, November 27, 2011

comparing R, Matlab, SAS, STATA, SPSS

The following table compares the standard procedures of the five packages in detail. By "standard" I mean built-in or readily available from the official or widely known and reliable public web-sites.
 Nonparametric Tests Yes Yes Yes Yes Yes
 T-test Yes Yes Yes Yes Yes
 ANOVA & MANOVA Yes Yes Yes Yes Yes
 ANCOVA & MANCOVA Yes Yes Yes Yes Yes
 Linear Regression Yes Yes Yes Yes Yes
 Generalized Least Squares Yes Yes Yes  Yes Yes 
 Ridge Regression Yes Yes Yes   
 Lasso Yes Yes Experimental   
 Generalized Linear Models Yes Yes Yes Yes Yes
 Mixed Effects Models Yes Yes Yes Yes Yes
 Logistic Regression Yes Yes Yes Yes Yes
 Nonlinear Regression Yes Yes Yes   
 Discriminant Analysis Yes Yes Yes  Yes  Yes 
 Nearest Neighbor Yes Yes Yes   Yes 
 Factor & Principal Components Analysis Yes Yes Yes Yes Yes
 Copula Models Yes Yes Experimental  
 Cross-Validation Yes Yes Yes   
 Bayesian Statistics Yes Yes Limited  
 Monte Carlo, Classic Methods Yes Yes Yes  Yes  Limited
 Markov Chain Monte Carlo Yes Yes Yes   
 Bootstrap & Jackknife Yes Yes Yes  Yes 
 EM Algorithm Yes Yes Yes   
 Missing Data Imputation Yes Yes Yes  Yes  Yes 
 Outlier Diagnostics Yes Yes Yes  Yes  Yes
 Robust Estimation Yes Yes Yes  Yes 
 Longitudinal (Panel) Data Yes Yes Yes  Yes 
 Survival Analysis Yes Yes Yes  Yes  Yes 
 Path Analysis Yes Yes Yes   
 Propensity Score Matching Yes Yes Limited   
 Stratified Samples (Survey Data) Yes Yes Yes  Yes  Yes 
 Experimental Design Yes Yes   
 Quality Control Yes Yes  Yes  Yes 
 Reliability Theory Yes Yes Yes  Yes  Yes
 Univariate Time Series Yes Yes Yes  Yes  Limited
 Multivariate Time Series Yes Yes Yes  Yes  
 Markov Chains Yes Yes   
 Hidden Markov Models Yes Yes   
 Stochastic Volatility Models Yes Yes Limited Limited  Limited 
 Diffusions Yes Yes   
 Counting Processes Yes Yes Yes   
 Filtering Yes Yes Limited  Limited 
 Instrumental Variables Yes Yes Yes Yes 
 Simultaneous Equations Yes Yes Yes  Yes 
 Splines Yes Yes Yes   
 Nonparametric Smoothing Methods Yes Yes Yes  Yes  
 Extreme Value Theory Yes Yes   
 Variance Stabilization Yes Yes   
 Cluster Analysis Yes Yes Yes  Yes  Yes 
 Neural Networks Yes Yes Yes   Limited
 Classification & Regression Trees Yes Yes Yes   Limited
 Boosting Classification & Regression Trees Yes Yes   
 Random Forests Yes Yes   
 Support Vector Machines Yes Yes   
 Signal Processing Yes Yes   
 Wavelet Analysis Yes Yes Yes  
 ROC Curves Yes Yes Yes  Yes  Yes 
 Optimization Yes Yes Yes   

Please read the detailed description of the services offered in the areas of statistical consulting and financial consulting: home pagetypes of serviceexperiencecase studies and payment options. You may also find the following pages useful: statistics resources and finance resources.