python笔记：4.2.1统计表的基本要素

# -*- coding: utf-8 -*-
"""
Created on Sat Jun  1 15:28:56 2019

@author: User
"""


import pandas as pd
import numpy as np
from scipy import stats

stock=np.dtype([('id',np.str,5),
                ('time',np.str,10),
                ('code',np.str,10),
                ('open_p',np.float64),
                ('close_p',np.float64),
                ('low_p',np.float64),
                ('vol',np.int32),
                ('high_p',np.float64),
                ('col',np.int32)])

print('\n jd_stock:')
jd_stock=np.loadtxt('data\stock.csv',delimiter=',',dtype=stock)
print(jd_stock)

print('\n jddf:')
jddf=pd.read_csv('data\stock.csv',sep=',',header=None,
                   names=['id','time','code','open_p','colse_p','low_p','vol','high_p','col'])
print(jddf.head())

#print("\n :")
#print(jddf['open_p'].skew())
print("以上是数据准备----------------------------")

print("\n  各种变量的简单统计描述 jddf.describe():")
print(jddf.describe())

print("\n  用 scipy 进行各种变量的简单统计描述:")
print(stats.describe(jddf[['open_p','colse_p']]))

输出：

 jd_stock:
[('1', '20130902', '600028', 4.41, 4.43, 4.37,  17275, 4.41,  392662)
 ('2', '20130903', '600028', 4.41, 4.46, 4.4 ,  19241, 4.45,  434177)
 ('3', '20130904', '600028', 4.44, 4.49, 4.42,  20106, 4.47,  451470) ...
 ('1356', '20190327', '600028', 5.71, 5.75, 5.69,  63601, 5.72, 1112544)
 ('1357', '20190328', '600028', 5.69, 5.7 , 5.62,  65692, 5.64, 1162484)
 ('1358', '20190329', '600028', 5.65, 5.75, 5.61, 112785, 5.74, 1981482)]

 jddf:
   id      time    code  open_p  colse_p  low_p       vol  high_p     col
0   1  20130902  600028    4.41     4.43   4.37  17275.39    4.41  392662
1   2  20130903  600028    4.41     4.46   4.40  19241.84    4.45  434177
2   3  20130904  600028    4.44     4.49   4.42  20106.30    4.47  451470
3   4  20130905  600028    4.47     4.48   4.42  15582.48    4.47  349997
4   5  20130906  600028    4.46     4.52   4.45  19101.41    4.50  425777
以上是数据准备----------------------------

  各种变量的简单统计描述 jddf.describe():
                id          time  ...       high_p           col
count  1358.000000  1.358000e+03  ...  1358.000000  1.358000e+03
mean    679.500000  2.016016e+07  ...     5.668733  1.862497e+06
std     392.165144  1.638818e+04  ...     0.838200  2.301859e+06
min       1.000000  2.013090e+07  ...     4.270000 -1.656342e+07
25%     340.250000  2.015012e+07  ...     4.990000  6.711858e+05
50%     679.500000  2.016062e+07  ...     5.675000  1.063988e+06
75%    1018.750000  2.017111e+07  ...     6.190000  1.921891e+06
max    1358.000000  2.019033e+07  ...     8.670000  1.645562e+07

[8 rows x 9 columns]

  用 scipy 进行各种变量的简单统计描述:
DescribeResult(nobs=1358, minmax=(array([4.24, 4.29]), array([8.7 , 9.27])), mean=array([5.6587187 , 5.73196613]), variance=array([0.69569784, 0.74655759]), skewness=array([0.52696423, 0.58782836]), kurtosis=array([-0.29442338, -0.16281185]))