R/data.norm.R

In jservadio/TrialistNof1: Bayesian N-of-1 calculations

data.norm <- function (Y, Covs, prior, Treat, mod.id) 
{
    inData = list(Treat = Treat, Y = Y, mean.alpha = prior$mean.alpha, prec.alpha = prior$prec.alpha, 
		mean.beta = prior$mean.beta, prec.beta = prior$prec.beta, prec.1 = prior$prec.1, prec.2 = prior$prec.2)
    if (!is.null(Covs)) {
		if (mod.id != 4) {
			inData[[1 + length(inData)]] = Covs
			names(inData)[[length(inData)]] = "x"
		}
        inData[[1 + length(inData)]] = prior$mean.slope
        names(inData)[[length(inData)]] = "mean.slope"
        inData[[1 + length(inData)]] = prior$prec.slope
        names(inData)[[length(inData)]] = "prec.slope"
		
		if (mod.id == 2 || mod.id == 4.1) { tm = ncol(inData$x)
		} else if (mod.id == 5.2) { tm = 1 
		} else if (mod.id == 5.41) { tm = 2 }
		
		if (mod.id %in% c(2, 4.1, 5.2, 5.41)) {
			x = inData$x[ , tm]
			j = l = x
			
			if (is.na(j[1])) { j[1] = 1 }
			if (is.na(l[length(l)])) { l[length(l)] = l[length(l) - 1] + 1 }
			for (ix in 2:length(x)) {
				if (is.na(j[ix])) { j[ix] = j[ix - 1] }
			}
			for (ix in (length(x) - 1):1) {
				if (is.na(l[ix])) { l[ix] = l[ix + 1] }
			}
			for (i in (length(j) - 1):1) {
				if (is.na(j[i])) { j[i] = j[i + 1] - 1 }
			}
			for (i in 2:length(l)) {
				if (is.na(l[i])) { l[i] = l[i - 1] + 1 }
			}
			j = j - .05
			l = l + .05		

			inData[[length(inData) + 1]] = j
			names(inData)[[length(inData)]] = "j"
			inData[[length(inData) + 1]] = l
			names(inData)[[length(inData)]] = "l"
		}
		
		if (mod.id %in% c(4, 4.1, 4.2, 5.4, 5.41, 5.42)) {
			v = mean(Y[!is.na(Y)])
			
			inData[[length(inData) + 1]] = v
			names(inData)[[length(inData)]] = "v"
		}
    }
    return(inData)
}