暢銷書數量分析

場景：

• • 書店某季採購暢銷書紀錄如附件檔。

問題：

• • 求算不同暢銷書數量分佈（Frequency Distribution）

  • 求算不同暢銷書百分比分佈（Percentage Frequency Distribution）

GNU R：

source("Statistics.R") ######################################################## # 4 # 已知 myData <- scan(file="Descriptive-Statistics-6.dat", what=character(0), sep="\n", multi.line=TRUE) myBooks <- sort(unique(myData)) # 求算 myTotalCount <- length(myData) myDataStat <- c(1:length(myBooks)) names(myDataStat) <- myBooks for (myBookInStore in myData) { for (myBookNum in c(1:length(myBooks))) { if (myBookInStore == myBooks[myBookNum]) { myDataStat[myBookNum] <- myDataStat[myBookNum] + 1 break } } } print(myDataStat) print(summary(myDataStat)) myDataFrame <- data.frame(myBooks, myDataStat, row.names=NULL) names(myDataFrame)[1] <- "暢銷書" names(myDataFrame)[2] <- "書本數" print(myDataFrame) myFrequency <- myDataStat / myTotalCount myDataFrame <- data.frame(myBooks, myFrequency, row.names=NULL) names(myDataFrame)[1] <- "暢銷書" names(myDataFrame)[2] <- "書本數相對頻率" print(myDataFrame) myPercentageFrequency <- 100 * myFrequency myDataFrame <- data.frame(myBooks, myPercentageFrequency, row.names=NULL) names(myDataFrame)[1] <- "暢銷書" names(myDataFrame)[2] <- "書本數頻率 %" print(myDataFrame) par(mfrow=c(3,2)) pie(myDataStat, main="暢銷書書本數餅圖", labels=names(myDataStat)) dotchart(myDataStat, main="暢銷書書本數點圖", xlab="書本數") barplot(myDataStat, main="暢銷書書本數長條圖", xlab="暢銷書", ylab="書本數", names.arg=myBooks) hist(myDataStat, main="暢銷書書本數直方圖", xlab="暢銷書", ylab="書本數", nclass=5) boxplot(myDataStat, main="暢銷書書本數盒狀圖", horizontal=TRUE)

解答：