Marketing Analytics (R)

x<-read.xlsx(file="C:\\Users\\Scott\\Desktop\\cereal.xls", sheetIndex=1,

startRow = 1, endRow = 44, header = TRUE, colIndex = 1:10,

encoding = "UTF-8")

#### Create the data

# Names of the variables we will define for each segment

segVars <- c("age", "gender", "income", "kids", "ownHome", "subscribe")

# the data type for each segment

segVarType <- c("norm", "binom", "norm", "pois", "binom", "binom")

# names of the segments

segNames <- c("Suburb mix", "Urban hip", "Travelers", "Moving up")

# the size of each segment (N)

segSize <- c(100, 50, 80, 70)

# the means for each variable for each segment

segMeans <- matrix( c(

40, .5, 55000, 2, .5, .1,

24, .7, 21000, 1, .2, .2,

58, .5, 64000, 0, .7, .05,

36, .3, 52000, 2, .3, .2 ), ncol=length(segVars), byrow=TRUE)

# the standard deviations for each segment (NA = not applicable for the variable)

segSDs <- matrix( c(

5, NA, 12000, NA, NA, NA,

2, NA, 5000, NA, NA, NA,

8, NA, 21000, NA, NA, NA,

4, NA, 10000, NA, NA, NA ), ncol=length(segVars), byrow=TRUE)

# make sure we're starting our dataset from a known state

seg.df <- NULL

set.seed(02554)

# iterate over all the segments and create data for each

for (i in seq_along(segNames)) {

cat(i, segNames[i], "\n")

# create an empty matrix to hold this particular segment's data

this.seg <- data.frame(matrix(NA, nrow=segSize[i], ncol=length(segVars)))

# within a segment, iterate over the variables and draw appropriate random data

for (j in seq_along(segVars)) { # and iterate over each variable

if (segVarType[j] == "norm") { # draw random normals

this.seg[, j] <- rnorm(segSize[i], mean=segMeans[i, j], sd=segSDs[i, j])

} else if (segVarType[j] == "pois") { # draw counts

this.seg[, j] <- rpois(segSize[i], lambda=segMeans[i, j])

} else if (segVarType[j] == "binom") { # draw binomials

this.seg[, j] <- rbinom(segSize[i], size=1, prob=segMeans[i, j])

} else {

stop("Bad segment data type: ", segVarType[j])

}

}

# add this segment to the total dataset

seg.df <- rbind(seg.df, this.seg)

}

# make the data frame names match what we defined

names(seg.df) <- segVars

# add segment membership for each row

seg.df$Segment <- factor(rep(segNames, times=segSize))

# convert the binomial variables to nicely labeled factors

seg.df$ownHome <- factor(seg.df$ownHome, labels=c("ownNo", "ownYes"))

seg.df$gender <- factor(seg.df$gender, labels=c("Female", "Male"))

seg.df$subscribe <- factor(seg.df$subscribe, labels=c("subNo", "subYes"))

# check the data and confirm it

summary(seg.df)