A new package : Quandl

Quandl is a new database management tool which seeks to become the place to find datasets. That is, each unique indicator is considered an independent data set. This helps them to seem to have a ginormous quantity of data sets. Source : Blog Econometric Simulation.
To load or find the datasets, we have to authentify using the API like with Twitter First, we need to set an account to receive the Quandl.auth

setwd("D:/PERSO/R_Working/Tutoriels/Probability")
library(Quandl)
Quandl.auth("XXXXXXXXXX")  ### Replace with yours

For example, we load the database about pollution and gdp and try to find the link.

plot(date, pollution, col = "red", type = "o", lwd = 2, ylim = c(70, 150), ylab = "", 
    main = "Evolution du PIB vs Pollution entre 94 et 2006 au Japon")
lines(date, pib, lty = 2, col = "purple", type = "o")
legend("topright", legend = c("PIB", "Pollution"), col = c("purple", "red"), 
    pch = 15, bty = "n", pt.cex = 2, cex = 0.8, text.col = "black", horiz = TRUE, 
    inset = c(0.1, 0.1))

We fit a model to know if the growth of GDP is responsible of pollution's growth in japan between 1994 and 2006.

Ecologists are not totaly wrong ! ! !

The entire code to run this post

	lm_eqn = function(df) {
	m = lm(pollution ~ pib, df);
	l <- list(a = format(coef(m)[1], digits = 2),
	b = format(abs(coef(m)[2]), digits = 2),
	r2 = format(summary(m)$r.squared, digits = 3));
	if (coef(m)[2] >= 0) {
	eq <- substitute(italic(Pollution) == a + b %.% italic(PIB)*","~~italic(R)^2~"="~r2,l)
	} else {
	eq <- substitute(italic(Pollution) == a - b %.% italic(PIB)*","~~italic(R)^2~"="~r2,l)
	}
	as.character(as.expression(eq));
	}
	require(ggplot2)
	p=ggplot(dd, aes(x=pib, y=pollution)) +geom_point(shape=5) ;
	p1 = p + geom_text(aes(x =110, y = 110, label = lm_eqn(dd)), parse = TRUE)+
	geom_smooth(method=lm)
	print(p1)

view raw quandl.r hosted with ❤ by GitHub

Share it ! ! !