博客
关于我
强烈建议你试试无所不能的chatGPT,快点击我
Excel2007制作直方图和正态分布曲线图
阅读量:5025 次
发布时间:2019-06-12

本文共 2590 字,大约阅读时间需要 8 分钟。

对同一维度的数据分析数据分布范围及分布趋势,要通过制作直方图和正态分布曲线图体现。

例如:已知所有员工的日收入,分析员工收入分布情况(51.7,50.6,57.9,56.9,56.7,56.7,55.3,56.1,53.7,54.5,56.9,51.9,52.1,55.1,54.9,54.7,55.3,55.3,54.5,54.9,54.5,55.3,54.9,54.3,53.7,53.5,53.7,53.1,54.5,53.1,53.9,53.5,53.3,53.9,53.5,53.5,52.5,53.3,53.5,53.3,53.7,53.1,54.5,53.9,56.7,54.5,54.3,55.1,54.1,54.5,53.9,53.1,53.3,55.3,55.7,56.1,54.7,53.1,53.3,52.7,53.1,52.9,53.1,54.3,53.1,52.7,53.1,53.3,53.1,53.3,53.1,53.3,55.1,54.7,54.9,54.3,53.9,53.7,53.9,53.5,54.5,54.3,55.5,55.7,55.5,54.9,55.3,55.5,53.7,54.1,53.9,55.7,55.9,53.7,53.5,53.1,52.3,52.7,52.9,53.3,53.9,52.7,53.5,53.1,52.7,51.9,52.5,53.9,54.5,55.7,55.3,54.9,53.1,52.9,54.1,53.3,54.7,53.9,54.3,54.1,53.7,53.3,52.7,52.9,52.5,53.9,53.5,54.1,54.1,54.7,54.9,54.9,54.1,53.3,52.9,53.7,53.9,54.3,54.1,54.5,54.7,54.9,52.1,52.9,53.5,52.7,53.1,53.1,53.5,52.9,52.9,53.1,53.3,52.7,53.5,53.9,54.9,55.1,54.3,55.1,54.3,54.3,53.9,54.5,54.5,54.3,55.3,54.5,54.9,53.5,52.1,55.3,55.7,55.7,55.5,54.5,57.7,54.7,53.7,53.1,53.7,55.9,56.1,53.9,53.7,53.3,53.9,53.9,54.5,54.7,56.1,55.7,53.1,53.7,53.5,53.9,53.9,53.5,53.3,53.1,52.5,55.9,55.7,54.1,54.3,54.1,54.1,54.5,54.5,55.1,53.1,53.3,54.1,54.3,53.9,54.1,54.7,54.7,53.7,53.1,53.3,52.7,53.5,52.9,53.7,56.5,56.1,55.7,55.5,56.9,57.7,56.5,55.7,54.1,54.7,55.7,55.5,53.1,52.7,53.1,53.3,53.5,54.3,54.1,54.5,54.7,55.7,55.5,54.1,54.3,54.7,53.1,53.3,53.1,52.7,53.1,53.7,53.1,54.7,54.5,55.1,54.7,54.5,56.1,55.7,53.3,52.5,53.7,54.1,53.3,52.1,52.3,53.1,53.3,53.5,53.3,53.1,52.7,53.1,55.7,55.1,54.3,53.7,53.1,52.9,53.1,52.7,52.5,53.1,53.5,53.1,53.3,54.1,55.1,54.9,56.1,55.7,56.5,54.7,53.7)

    一、制作直方图

  • 将数据输入到EXCEL同一列中(这里放入A列);
  • 计算“最大值”、“最小值”、“极差”、“分组数”、“分组组距”;

最大值:max(A:A);(=57.9)

最小值:min(A:A);(=50.6)

极差:最大值-最小值;(=7.3)

分组数:roundup(sqrt(count(A;A)),0);(=18)/*count(A:A)计算A列包含数字的单元格个数,sqrt求平方根,roundup按指定位数对数据进行向上四舍五入*/;

分组组距:极差/分组数;(0.4)

  • 数据分组:选一个比最小值小的一个恰当的值作为第一个组的起始坐标,然后依次加上“分组组距”,直到最后一个数据值比“最大值”大为止。

这里第一个组的起始坐标选为50.5,依次增加0.4,最后一组坐标为58.2,共计20组

  • 统计频率:统计每个分组中所包含的数据的个数。

方法:采用FREQUENCY函数,以一列垂直数组返回一组数据的频率分布,

1、=frequency(原始数据的范围,直方图分组的数据源);

2、先选中将要统计直方图每个子组中数据数量的区域

3、再按“F2”健,进入到“编辑”状态

4、再同时按住“Ctrl”和“Shift”两个键,再按“回车Enter”键,最后三键同时松开.

  • 制作直方图:选择频率数插入柱状图
  • 修整柱形图:设置数据系列格式-调制无间距

    二、制作正态分布图

  • 获取正态分布概念密度:NORMDIST(作用:返回指定平均值和标准偏差的正态分布函数)

语法:

NORMDIST(x,mean,standard_dev,cumulative)

X 为需要计算其分布的数值;(以每一个分组边界值为“X”,依次往下拉)

Mean 分布的算术平均值;(Mean=AVERAGE(A:A)(数据算术平均))【这里为54.09】

Standard_dev 分布的标准偏差;(Standard_dev=STDEV.S(A:A)(数据的标准方差)【1.15】

Cumulative=false(概率密度函数)

Cumulative 为一逻辑值,指明函数的形式。如果 cumulative 为 TRUE,函数 NORMDIST 返回累计分布函数;如果为 FALSE,返回概率密度函数。

  • 在直方图中增加正态分布曲线图:设置曲线图,选择次坐标轴

 

转载于:https://www.cnblogs.com/smallcrystal/p/4910009.html

你可能感兴趣的文章
Android Studio默认快捷键
查看>>
发布开源库到JCenter所遇到的一些问题记录
查看>>
第七周作业
查看>>
函数式编程与参数
查看>>
flush caches
查看>>
SSAS使用MDX生成脱机的多维数据集CUB文件
查看>>
ACM_hdu1102最小生成树练习
查看>>
MyBatis源码分析(一)--SqlSessionFactory的生成
查看>>
android中ListView点击和里边按钮或ImageView点击不能同时生效问题解决
查看>>
CTF常用工具之汇总
查看>>
java的面向对象 (2013-09-30-163写的日志迁移
查看>>
HDU 2191 【多重背包】
查看>>
51nod 1433 0和5【数论/九余定理】
查看>>
【AHOI2013复仇】从一道题来看DFS及其优化的一般步骤和数组分层问题【转】
查看>>
less 分页显示文件内容
查看>>
如何对数据按某列进行分层处理
查看>>
[Qt] this application failed to start because it could not find or load the Qt platform plugin
查看>>
Git Submodule管理项目子模块
查看>>
学会和同事相处的30原则
查看>>
NOJ——1568走走走走走啊走(超级入门DP)
查看>>