3.18 Chapter 1
Histogram (柱状图)
一、Key Concepts 几个概念
1.Population (总数)：a population is the group of all items of interest to a statistics practitioner.
2.Sample (样本) ：A sample is a set of data drawn from the population.
[Part of a population]
3.Parameter(参数): A descriptive measure of apopulation.
4.Statistic: A descriptive measure of a sample.
二、Statistics 分类
1.Descriptive Statistics（描述性统计）：organizing, summarizing,and presenting data.
Include: Graphical Techniques&Numerical Techniques
2.Inferential Statistics （推断统计）：draw conclusionsor inferences about characteristics of populations based on data from a sample.
三、Significance level（显著性水平）和Confidence level （置信水平）
1.The confidence level is the proportion of times that an interval estimate for a population parameter will be correct.“ 1–α” torepresent the confidence level when we wish to estimate a population parameter.
2.Significance level measures how frequently a “true claim” is accidently rejected. Use α(Greek letter “alpha”) tore present the significance level when testing a claim about a population parameter .
https://www.zhihu.com/question/23149768?utm_campaign=rss&utm_medium=rss&utm_source=rss&utm_content=title
Chapter 2 Numerical DescriptiveTechniques数值描述方法
第一部分
A variable（变量）[Typicallycalled a “random” variable since we do not know it’s value until we observe it]is some characteristic of a population or sample.
The values of the variable are the range of possible values for avariable.
一、Two types of Data
1.Numerical/Quantitative Data [Real Numbers] -定量数据 Include Inteval data & Ratio data. 包括定距数据和定比数据
a.Continuous Data – Data can be any real number within a given range. 连续性数据
b.Discrete Data – Data can only be very specific 离散型数据
values which we can list.
2.Qualitative/Categorical Data [Labels rather than numbers]-定性数据
包括定类数据和定序数据
a.Frequency
distribution：the data in a tablethat presents the categories and their counts
b.Relative frequency distribution: liststhe categories and the proportion with which each occurs.
c.Pareto chart(帕累托表) : Since Nominal data has no order, if we
arrange the outcomes from the most frequently occurring to the least frequently occurring
d.Cross-sectional data: Observations
measured at the same point in time
e.Time-series data: Observations measured at successive points in time
计量经济学的两大研究对象：横截面数据（Cross-sectional
Data）和时间序列数据（Time-series Data）。前者旨在归纳不同经济行为者是否具有相似的行为关联性，以模型参数估计结果显现相关性；后者重点在分析同一经济行为者不同时间的资料，以展现研究对象的动态行为。
Line Cart （线性图）
第二部分
Numerical Descriptive Techniques...
1.Measures of Central Location（中量数）：Mean（平均数）, Median（中间的数）, Mode（众数）
2.Measures of Variability（差异量数）：Range（范围）, Standard Deviation（标准差）, Variance（方差）, Coefficient of Variation（变异系数）
3.Measures of Relative Standing（相对位置量数） Percentiles（百分位数）, Quartiles （四分位数）
4.Measures of Linear Relationship Covariance（线性关系协方差测度
）Correlation（相关系数）, Least Squares Line（最小二乘直线）
arithmetic mean 算数平均数
The Empirical Rule（经验法则）： 经验规则是统计规律，指出了在正态分布，几乎所有数据都将落在均值的三倍标准差内。所述经验规则表明，68%的数据将分布在的第一个标准偏差之内，95%，和99.7%将落在均值的前三个标准偏差之内。
Box Plots...（箱体图）：箱体图是在1977年由美国的统计学家约翰·图基(John Tukey)发明的。它由五个数值点组成：最小值(min)，下四分位数(Q1)，中位数(median)，上四分位数(Q3)，最大值(max)。
http://www.blogjava.net/norvid/articles/317235.html