core

Main functions of estout package

Short descriptions for main functions (see below for more details):

Exported source 
from __future__ import annotations
from pathlib import Path 
from typing import List, Dict, Literal, Union, Callable
import importlib

import numpy as np
import pandas as pd
import statsmodels.api as sm
from linearmodels import PanelOLS

from estout import utils

Set up an example dataset and run a few regressions to showcase the functions in this module.

np.random.seed(123)
df = pd.DataFrame(np.random.rand(9,3), 
                  columns=['y','x','z'],
                  index = pd.MultiIndex.from_product([[1,2,3],[1,2,3]], names=['firmid','time'])
                  ).assign(cons = 1)
sm1 = sm.OLS(df['y'], df[['cons','x']]).fit()
sm2 = sm.OLS(df['y'], df[['cons','x','z']]).fit().get_robustcov_results(cov_type='HAC', maxlags=2)
lmres = PanelOLS(df['y'],  df[['cons','x','z']], entity_effects=True
                 ).fit(cov_type='clustered', cluster_entity=True)

source

collect_stats

 collect_stats (res, get_default_stats=True, add_stats:dict=None,
                add_literals:dict=None)

Collects stats from res object. Stats in add_stats can override default stats

Type Default Details
res Results object to extract stats from
get_default_stats bool True If True, returns all stats implemented by the f'{package}_results' module
add_stats dict None Keys are stats to extract in addition to the default ones; values are attributes of res or callables
add_literals dict None Additional info to be added to output dict; values must be scalars
Returns dict 
stats1 = collect_stats(sm1)
stats2 = collect_stats(sm2, add_literals={'Cov Type': 'Newey West'})
stats3 = collect_stats(lmres, add_stats={'r2b': 'rsquared_between', 'FE':'included_effects'})
stats4 = collect_stats(sm2, add_stats={'Max Eigen': lambda x: x.eigenvals.max()})
stats1
{'package': 'statsmodels',
 'ynames': ['y'],
 'xnames': ['cons', 'x'],
 'params': cons    0.507852
 x       0.345003
 dtype: float64,
 'tstats': cons    3.905440
 x       1.292246
 dtype: float64,
 'pvalues': cons    0.005858
 x       0.237293
 dtype: float64,
 'covmat':           cons         x
 cons  0.016910 -0.030531
 x    -0.030531  0.071278,
 'se': cons    0.130037
 x       0.266979
 dtype: float64,
 'nobs': 9,
 'r2': np.float64(0.19260886185799475)}
assert stats2['Cov Type'] == 'Newey West'
stats3['r2b']
0.7954933715233719
stats3['FE']
['Entity']

source

to_df

 to_df (res_list:List[dict], which_xvars:list=None,
        stats_body:list=['params', 'tstats'], stats_bottom:list=['r2',
        'nobs'], labels:dict=None, add_formats:dict=None)

Combines results from multiple collect_stats() outputs into a single pd.DataFrame

Type Default Details
res_list List List of outputs from collect_stats()
which_xvars list None If None, report all xvars
stats_body list [‘params’, ‘tstats’] Each element of res_list needs to have these stats as keys; values must be pd.Series
stats_bottom list [‘r2’, ‘nobs’] Each element of res_list needs to have these stats as keys; values must be scalars
labels dict None
add_formats dict None
Returns DataFrame 
d = to_df(res_list=[stats1, stats2, stats1, stats3], 
          which_xvars=['cons','x','z'], 
          add_formats={'r2':'{:.2f}'},
          labels={'nobs':'Observations', 'r2':'$R^2$'})
d
0 1 2 3
cons params 0.51*** 0.70*** 0.51*** 0.73***
tstats (3.91) (21.48) (3.91) (167.36)
x params 0.35 0.57** 0.35 0.64*
tstats (1.29) (2.85) (1.29) (2.26)
z params -0.64** -0.77**
tstats (-3.55) (-2.91)
$R^2$ 0.19 0.49 0.19 0.35
Observations 9 9 9 9

Note how we used a LaTex-friendly label for the \(R^2\) coefficient so we don’t have to re-label this row before we send it to LaTex.

source

to_tex

 to_tex (dfs:Union[pandas.core.frame.DataFrame,List[pandas.core.frame.Data
         Frame]], outfile:pathlib.Path|str=None, title:str='Table title',
         notes:str='Table description', notes_on_top:bool=True,
         label:str='',
         table_type:Literal['table','sidewaystable']='table',
         font_size:str='\\footnotesize', addtocounter:int=0,
         panel_title:List[str]=None, palign:Literal['l','r','c']='l',
         col_groups:List[dict]=None,
         col_names:List[Union[list,bool]]=True,
         hlines:List[List[int]]=None, tabular_env:str='tabular*')

Create tex code to generate table from one or more dataframes

Type Default Details
dfs Union DataFrame(s) to be converted to tex table; if multiple, they will be panels in a larger table
outfile pathlib.Path | str None Where to save resulting tex output
title str Table title Table title
notes str Table description Some call this the table caption
notes_on_top bool True Set to False if you want table description (caption) to be at the bottom
label str Table label (for referencing within LaTex document)
table_type Literal table
font_size str Gets applied to the table contents as well as its caption
addtocounter int 0 Set to -1 for tables that are just a continuation of a table on a new page
panel_title List None One element in the list for each dataframe in dfs
palign Literal l Alignment of panel title
col_groups List None Keys are group names; values are lists of consecutive indices of columns in the group
col_names List True If False, none; if True, use df column names; if list, gives custom column names
hlines List None Row indices under which to place hline
tabular_env str tabular* LaTex tabular environment
Returns str 
tbl = to_tex([d,d], panel_title=['Panel A: Some title', 'Panel B: Some title'], 
               col_groups=[{'Group1':[1,2]}]*2,
               col_names=[['Model 1', 'Model 2', 'Model 3', 'Model 4']]*2,
               hlines=[[0,1,4,12], [1,4,12]] )

source

to_pdf

 to_pdf (outfile:str, table_tex_code:Union[str,pathlib.Path,List[Union[str
         ,pathlib.Path]]]=None,
         article_spec='\\documentclass[11pt]{article}',
         captionsetup='format=plain, labelsep=newline, labelfont = bf,
         justification=centering', make_pdf:bool=True,
         open_pdf:bool=False)

Creates PDF with one or more tables given their tex code (from to \end{table})

Type Default Details
outfile str Path to .tex file where combined tables are saved (must contain .tex extension)
table_tex_code Union None String(s) or Paths to files containing table tex code (e.g. like outputs of to_tex())
article_spec str
captionsetup str format=plain, labelsep=newline, labelfont = bf, justification=centering
make_pdf bool True
open_pdf bool False

For the test below to work, you need to have TexLive installed (and change the path below to a valid path on your system).

to_pdf('../_outputs/paper.tex', table_tex_code=[tbl, tbl])
PDF creation successful!