Stat Tools - Demosophy.org

to make loading of GUI automatic Navigate to R’s etc directory on your hard drive. It should be in a location such as: “C: \Program Files \R \R-2.15.0 \etc. In the
etc directory, there should be a file called Rprofile.site. Open this file using a text editor such as Notepad,
Wordpad, or Word. At the end of this file, you need to add the following lines, exactly as they are typed
below:
local({
old <- getOption(“defaultPackages”)
options(defaultPackages = c(old,”Rcmdr”))
})

local({
old <- getOption(“defaultPackages”)
options(defaultPackages = c(old,”rattle”))
})

————————–

to run JGR in R

If JGR doesn’t run in R

Install R packages and add Deducer

Install R packages rJava

Install R packages JGR

Make sure you have 64 bit Java from http://www.java.com/en/download/manual.jsp

Sys.setenv(JAVA_HOME=’C:\\Program Files\\Java\\jre1.8.0_25′)
library(rJava)

library(JGR)

JGR()

———-

To Have context sensitive IDE and two GUI applications (RKWard and Deducer)

RKWARD is an IDE GUI context sensitive (not very powerful in term of stat capabilities)

install rkward will install r

goto setting-manage R packages and add Deducer

goto setting-manage R packages and JGR and rJava

JGR()

In JGR package manager make the packages you need load as default

now you have a context sensitive two IDE

Everytime start Rkward and type

library(JGR)
JGR()

——-

Rstudio is NOT context sensitive

Inzight Initially designed for New Zealand high schools, iNZight now extends to multivariable graphics, time series and generalised linear models (allows for complex survey designs)

https://www.stat.auckland.ac.nz/~wild/iNZight/

=======================================================================

PPSP has a nice correlation matrix but deducer is better

====================================================================

3rd- PSPP does a beautiful Chi squared when you do cross tab

Age * Habit [count, row %, column %, total %].
	*Habit*
*Age*	Not	smoke	Total
Young	4.00	1.00	5.00
	80.00%	20.00%	100.00%
	80.00%	16.67%	45.45%
	36.36%	9.09%	45.45%
old	1.00	5.00	6.00
	16.67%	83.33%	100.00%
	20.00%	83.33%	54.55%
	9.09%	45.45%	54.55%
Total	5.00	6.00	11.00
	45.45%	54.55%	100.00%
	100.00%	100.00%	100.00%
	45.45%	54.55%	100.00%

Chi-square tests.
*Statistic*	*Value*	df	*Asymp. Sig. (2-tailed)*	*Exact Sig. (2-tailed)*	*Exact Sig. (1-tailed)*
Pearson Chi-Square	4.41	1	.04
Likelihood Ratio	4.75	1	.03
Fisher’s Exact Test				.08	.07
Continuity Correction	2.23	1	.14
N of Valid Cases	11

==============================================

SOFA Statistics (good chi square and correlation between two vars) it has Kruskal-Wallis H test and Mann Whitney U Test for ordinal data

http://www.sofastatistics.com/home.php

PSPP interesting Chi square analysis for agree and disagree + cluster analysis (no plots)

http://www.gnu.org/software/pspp/

==============================================

RKWard is a separate easy to use and easily extensible IDE/GUI for R.

http://sourceforge.net/apps/mediawiki/rkward/index.php?title=Main_Page

http://sourceforge.net/apps/mediawiki/rkward/index.php?title=Screenshots

=================================

To learn R programming and data science

https://github.com/swirldev/swirl_courses

Get R

Install Rstudio

install.packages(“swirl”)

library(“swirl”)

swirl()

==================================================

All Probability distributions

http://www.predictiveanalyticstoday.com/top-30-free-statistical-software/

http://zeligproject.org/about.html

http://freestatistics.info/?p=stat

http://r.iq.harvard.edu/docs/zelig.pd

=======================================================

Network visualization

http://www.yworks.com/products/yed/download

=======================================================

What statistics that is possible in .NET? ( regression,pvalue ? )

StatisticFormula.Correlation

https://msdn.microsoft.com/en-us/library/system.windows.forms.datavisualization.charting.statisticformula.correlation%28v=vs.110%29.aspx

————————-

Chart.DataManipulator.FinancialFormula

https://msdn.microsoft.com/en-us/library/dd456655%28v=vs.100%29.aspx

Examples: One chart can have many curves

Chart1.DataManipulator.FinancialFormula (FinancialFormula.Forecasting, “2,40,true,true”, “Series1:Y”, “Series2:Y,Series3:Y,Series3:Y2”);

Chart1.DataManipulator.FinancialFormula(FinancialFormula.Forecasting, “Linear,0,false,false”, Chart1.Series[“amir series 1”], Chart1.Series[“Trendline 1”]);
Chart1.DataManipulator.FinancialFormula(FinancialFormula.Forecasting, “Linear,0,false,false”, Chart1.Series[“amir series 2”], Chart1.Series[“Trendline 2”]);

———————–

Azure

https://msdn.microsoft.com/en-us/library/azure/dn905978.aspx

Chart.DataManipulator.FinancialFormula(FinancialFormula.Forecasting,“RegressionType,Period,ApproxError,ForecastError”,“Historical”,
“Forecast,UpperError,LowerError”)

Data mining tools

http://eric.univ-lyon2.fr/~ricco/tanagra/en/tanagra.html

http://orange.biolab.si/

http://www.cs.waikato.ac.nz/ml/weka/

http://alphaminer.software.informer.com/2.0/

https://rapidminer.com/academia/educational-program/

Decision trees

http://eric.univ-lyon2.fr/~ricco/sipina.html