在 pandas to_stata 中设置索引标签

Question

我正在尝试将 pandas DataFrame 保存到 Stata 格式的文件中。更具体地说，DataFrame的索引需要保存到索引所在列的header必须有一个spacific text：换句话说我需要设置一个索引标签。

Pandas' to_csv has an index_label option, but pandas' to_stata 函数没有 index_label 选项。

保存为 Stata 格式时如何设置索引标签？

Answer 1

具有已经具有索引名称的 DataFrame（案例 1）与事先未设置索引名称（）之间存在细微差别Case2).

案例 1：index_name 已经到位（使用.set_index）

import pandas as pd
#data
data = [['Eren Jaeger', 15,'Soldier' ] , ['Mikasa Ackerman', 14,'Soldier'], ['Armin Arlert', 14,'Soldier'],['Levi Ackerman', 30, 'Captain']]  
#creating DataFrame
df = pd.DataFrame(data, columns = ['Name', 'Age', 'Rank'])
#setting index_name based on a previous variable
df = df.set_index('Rank', drop=True)
#creating dta file (no need of .rename_axis(index='my_index'))
df.to_stata('stata_df_1.dta' )  

df
##                Name  Age
## Rank                         
## Soldier      Eren Jaeger   15
## Soldier  Mikasa Ackerman   14
## Soldier     Armin Arlert   14
## Captain    Levi Ackerman   30

情况 2：索引未命名（需要 .rename_axis(index='my_index')）

根据@QuangHoang 的评论，这是一种在没有预先命名的情况下设置索引名称的方法。

data = [['Eren Jaeger', 15] , ['Mikasa Ackerman', 14], ['Armin Arlert', 14],['Levi Ackerman', 30]]  
df = pd.DataFrame(data, columns = ['Name', 'Age'])

df
##              Name  Age
## 0      Eren Jaeger   15
## 1  Mikasa Ackerman   14
## 2     Armin Arlert   14
## 3    Levi Ackerman   30

#this will have a first variable with digits 1 to 4 called "index" (default)
df.to_stata('stata_df_no_name.dta' )  

#this will have a first variable with digits 1 to 4 called "my_index"
df.rename_axis(index='my_index').to_stata('stata_df_2.dta')