【利用Python进行数据分析8】Pandas入门(描述性统计的汇总与计算)

导包

import yfinance as yf
import pandas as pd
import numpy as np

1.描述性统计的汇总与计算

#1.描述性统计的汇总与计算
#数值型数据
df=pd.DataFrame([[1.4,np.nan],[7.1,-4.5],[np.nan,np.nan],[0.75,-1.3]],
                index=["a","b","c","d"],
                columns=["one","two"])
#print(df.sum())#输出列之和的Series
#print(df.sum(axis=1))#输出行之和的Series
#print(df.sum(axis=1,skipna=False))#不跳过NA的计算，即有NA则和直接返回NA
#print(df.mean(axis=1))#算平均数，要求至少有一个非NA，此时计算的平均数是剩余非NA数的平均值
#print(df.idxmax())#最大值索引
#print(df.cumsum())#按列逐行累计加和
#print(df.describe())#生成多个汇总统计
#print("---------------------")
#非数值型
obj=pd.Series(["a","a","b","c"]*4)
#print(obj.describe())

2.相关系数与协方差

#2.相关系数与协方差
symbols=['AAPL','GOOG','IBM','MSFT']
#data=yf.download(symbols,"2016-10-17","2016-10-21")#下载历史五日数据,要tz,先留着（

3.唯一值，计数和成员属性

#3.唯一值，计数和成员属性
obj2=pd.Series(["c","a","d","a","a","b","b","c","c"])
uniques=obj2.unique()#生成唯一值的数组形式
#print(uniques)#数值型可用sort排序
#print(obj2.value_counts())#统计频次,按频次降序排列生成Series
#print(pd.Series(obj2.to_numpy()).value_counts())#统计频次,按频次降序排列生成Numpy

#print(obj2[obj2.isin(["b","c"])])#过滤指定项子集
vals=pd.Series(["c","b","a","d"])
#print(pd.Index(vals).get_indexer(obj2))#Index.get_indexer方法,输出vals规定的索引array

data2=pd.DataFrame({"Qu1":(1,3,4,3,4),
                    "Qu2":(2,3,1,2,3),
                    "Qu3":(1,5,2,4,4)})

#print(data2["Qu1"].value_counts().sort_index())#取一列计数
#print(data2.value_counts())#每行当作元组