财务数据探索分析

# _*_coding:utf-8  _*_
# 对数据进行基本的探索
# 返回缺失值个数以及最大最小值
import pandas as pd


datafile = r'chargeSKDJApplyExport.xls' #原始数据，第一行为属性标签
resultfile = r'view.xls'   #数据探索结果表
data = pd.read_excel(datafile, skiprows=1)  #读取原始数据，指定UTF-8编码（需要用文本编辑器将数据转换为UTF-8编码）
data = pd.DataFrame(data)
# data['登记时间'] = pd.DatetimeIndex(data['登记时间'])  #将“登记时间”设为日期索引格式
# data = data.set_index('登记时间')

# data = data[['登记时间'], ['委托单位'], ['收款金额'], ['数据采集时间']]
# print(data)
# view = data.describe(percentiles=[], include='all')
# print(view)
view = data.describe(percentiles=[], include='all').T
# print(view)
view['null'] = len(data)-view['count']
view =view[['null', 'max', 'min']]
view.columns=[u'空值数', u'最大值', u'最小值']
# print(view)

view.to_excel(resultfile)