用R进行描述性统计

d = read.table('上证综指.txt', header = FALSE)print('上证综指...')print('描述性分析...')#Begin,High,Low,Close,Volume,Turnover#d[,2];d[,3];d[,4];d[,5];d[,6];d[,7]d1 = c(d[, 2])d2 = c(d[, 3])d3 = c(d[, 4])d4 = c(d[, 5])d5 = c(d[, 6])d6 = c(d[, 7])#样本数count1 = length(d1)count2 = length(d2)count3 = length(d3)count4 = length(d4)count5 = length(d5)count6 = length(d6)#包括6个项目,分别为min,下四分之一分位数,median,mean,上四分之一分位数,maxsd1 = summary(d1)sd2 = summary(d2)sd3 = summary(d3)sd4 = summary(d4)sd5 = summary(d5)sd6 = summary(d6)#极差,四分位极差,标准误,方差,标准差,变异系数,校正平方和,未校正平方和,峰度系数,偏度系数vd1 = c(sd1[6] - sd1[1], quantile(d1, 3/4) - quantile(d1, 1/4), sd(d1) / count1, var(d1), sd(d1), 100 * sd(d1) / sd1[4], sum((d1 - sd1[4]) ^ 2) , sum(d1 ^ 2), (count1 ^ 2) * sum((d1 - sd1[4]) ^ 3) / ((sd(d1) ^ 3) * (count1 - 1) * (count1 -2 )), count1 * (count1 + 1) * sum((d1 - sd1[4]) ^ 4)/((count1 - 1) * (count1 - 2) * (count1 - 3) * ((sd(d1)) ^ 4 ) ) - 3 * ((count1 - 1) ^ 2) /((count1 - 2 ) * (count1 - 3)))vd2 = c(sd2[6] - sd2[1], quantile(d2, 3/4) - quantile(d2, 1/4), sd(d2) / count2, var(d2), sd(d2), 100 * sd(d2) / sd2[4], sum((d2 - sd2[4]) ^ 2) , sum(d2 ^ 2), (count2 ^ 2) * sum((d2 - sd2[4]) ^ 3) / ((sd(d2) ^ 3) * (count2 - 1) * (count2 -2 )), count2 * (count2 + 1) * sum((d2 - sd2[4]) ^ 4)/((count2 - 1) * (count2 - 2) * (count2 - 3) * ((sd(d2)) ^ 4 ) ) - 3 * ((count2 - 1) ^ 2) /((count2 - 2 ) * (count2 - 3)))vd3 = c(sd3[6] - sd3[1], quantile(d3, 3/4) - quantile(d3, 1/4), sd(d3) / count3, var(d3), sd(d3), 100 * sd(d3) / sd3[4], sum((d3 - sd3[4]) ^ 2) , sum(d3 ^ 2), (count3 ^ 2) * sum((d3 - sd3[4]) ^ 3) / ((sd(d3) ^ 3) * (count3 - 1) * (count3 -2 )), count3 * (count3 + 1) * sum((d3 - sd3[4]) ^ 4)/((count3 - 1) * (count3 - 2) * (count3 - 3) * ((sd(d3)) ^ 4 ) ) - 3 * ((count3 - 1) ^ 2) /((count3 - 2 ) * (count3 - 3)))vd4 = c(sd4[6] - sd4[1], quantile(d4, 3/4) - quantile(d4, 1/4), sd(d4) / count4, var(d4), sd(d4), 100 * sd(d4) / sd4[4], sum((d4 - sd4[4]) ^ 2) , sum(d4 ^ 2), (count4 ^ 2) * sum((d4 - sd4[4]) ^ 3) / ((sd(d4) ^ 3) * (count4 - 1) * (count4 -2 )), count4 * (count4 + 1) * sum((d4 - sd4[4]) ^ 4)/((count4 - 1) * (count4 - 2) * (count4 - 3) * ((sd(d4)) ^ 4 ) ) - 3 * ((count4 - 1) ^ 2) /((count4 - 2 ) * (count4 - 3)))vd5 = c(sd5[6] - sd5[1], quantile(d5, 3/4) - quantile(d5, 1/4), sd(d5) / count5, var(d5), sd(d5), 100 * sd(d5) / sd5[4], sum((d5 - sd5[4]) ^ 2) , sum(d5 ^ 2), (count5 ^ 2) * sum((d5 - sd5[4]) ^ 3) / ((sd(d5) ^ 3) * (count5 - 1) * (count5 -2 )), count5 * (count5 + 1) * sum((d5 - sd5[4]) ^ 4)/((count5 - 1) * (count5 - 2) * (count5 - 3) * ((sd(d5)) ^ 4 ) ) - 3 * ((count5 - 1) ^ 2) /((count5 - 2 ) * (count5 - 3)))vd6 = c(sd6[6] - sd6[1], quantile(d6, 3/4) - quantile(d6, 1/4), sd(d6) / count6, var(d6), sd(d6), 100 * sd(d6) / sd6[4], sum((d6 - sd6[4]) ^ 2) , sum(d6 ^ 2), (count6 ^ 2) * sum((d6 - sd6[4]) ^ 3) / ((sd(d6) ^ 3) * (count6 - 1) * (count6 -2 )), count6 * (count6 + 1) * sum((d6 - sd6[4]) ^ 4)/((count6 - 1) * (count6 - 2) * (count6 - 3) * ((sd(d6)) ^ 4 ) ) - 3 * ((count6 - 1) ^ 2) /((count6 - 2 ) * (count6 - 3)))#原始数据oData = matrix(c(d1, d2, d3, d4, d5, d6), byrow = FALSE, ncol = 6)#print(oData)#描述性统计head = c('Title', 'Open', 'High', 'Low', 'Close', 'Volume', 'Turnover')row1 = c('样本数', count1, count2, count3, count4, count5, count6)row2 = c('最小值', sd1[1], sd2[1], sd3[1], sd4[1], sd5[1], sd6[1])row3 = c('下四分之一分位数', sd1[2], sd2[2], sd3[2], sd4[2], sd5[2], sd6[2])row4 = c('中位数', sd1[3], sd2[3], sd3[3], sd4[3], sd5[3], sd6[3])row5 = c('均值', sd1[4], sd2[4], sd3[4], sd4[4], sd5[4], sd6[4])row6 = c('上四分之一分位数', sd1[5], sd2[5], sd3[5], sd4[5], sd5[5], sd6[5])row7 = c('最大值', sd1[6], sd2[6], sd3[6], sd4[6], sd5[6], sd6[6])row8 = c('极差', vd1[1], vd2[1], vd3[1], vd4[1], vd5[1], vd6[1])row9 = c('半极差', vd1[2], vd2[2], vd3[2], vd4[2], vd5[2], vd6[2])row10 = c('标准误', vd1[3], vd2[3], vd3[3], vd4[3], vd5[3], vd6[3])row11 = c('方差', vd1[4], vd2[4], vd3[4], vd4[4], vd5[4], vd6[4])row12 = c('标准差', vd1[5], vd2[5], vd3[5], vd4[5], vd5[5], vd6[5])row13 = c('变异系数', vd1[6], vd2[6], vd3[6], vd4[6], vd5[6], vd6[6])row14 = c('校正平方和', vd1[7], vd2[7], vd3[7], vd4[7], vd5[7], vd6[7])row15 = c('未校正平方和', vd1[8], vd2[8], vd3[8], vd4[8], vd5[8], vd6[8])row16 = c('峰度', vd1[9], vd2[9], vd3[9], vd4[9], vd5[9], vd6[9])row17 = c('偏度', vd1[10], vd2[10], vd3[10], vd4[10], vd5[10], vd6[10])result = matrix(c(head, row1, row2, row3, row4, row5, row6, row7, row8, row9, row10, row11, row12, row13, row14, row15, row16, row17), ncol = 7, byrow = TRUE)print(t(result))