betacon<-function(data){
	ncols <- length(seq(0.1, 0.9, 0.01))
	nrows <- length(seq(0.1, 1, 0.01))
	mu <- matrix(seq(0.1, 0.9, 0.01), nrow = nrows, ncol = ncols, byrow = T
		)
	theta <- matrix(seq(0.1, 1, 0.01), nrow = nrows, ncol = ncols, byrow = 
		F)	
# mu and theta are now matrices
# of the same dimensions : the
# rows of mu are copies of the
# vector (0.1, 0.11,...,0.9), and
# the columns of theta are copies
# of the vector (0, 0.1,...,1)'.
	loglik <- log(mu) * data[1]
	p <- mu
	s <- p
	for(i in 2:12) {
		p <- p * (1 - (mu + theta)/(1 + (i - 1) * theta))
		s <- s + p	# recursive specification of the
		loglik <- loglik + log(p) * data[i]
	# beta-geometric probabilities
	}
	p <- mu/mu - s
	loglik <- loglik + log(p) * data[13]
# completes log-likelihood
	 contour(mu[1,], theta[,1], t(loglik), nlevels = 200, xlab = "", ylab = "")
         persp(mu[1,], theta[,1], t(loglik),theta = 30, phi = 30, expand = 0.5, col = "lightblue", xlab = "mu", ylab = "theta", zlab = " ")
}
	data <- c(29, 16, 17, 4, 3, 9, 4, 5, 1, 1, 1, 3, 7)
betacon(data)
wmsmoke.fit<-fit.betageo(c(0.1,0.1),data)