Regression tables with huxreg

From version 0.2, huxtable includes the function huxreg to build a table of regressions.

huxreg can be called with a list of models. These models can be of any class which has a tidy method defined in the broom package. The method should return a list of regression coefficients with names term, estimate, std.error and p.value. That covers most standard regression packages.

Let’s start by running some regressions to predict a diamond’s price.

data(diamonds, package = 'ggplot2')

lm1 <- lm(price ~ carat + depth, diamonds)
lm2 <- lm(price ~ depth + factor(color, ordered = FALSE), diamonds)
lm3 <- lm(log(price) ~ carat + depth, diamonds)

Now, we call huxreg to display the regression output side by side.

huxreg(lm1, lm2, lm3)
(1) (2) (3)
(Intercept) 4045.333 *** 6491.466 *** 7.313 ***
(286.205)    (730.537)    (0.074)   
carat 7765.141 ***          1.971 ***
(14.009)             (0.004)   
depth -102.165 *** -53.835 *** -0.018 ***
(4.635)    (11.815)    (0.001)   
factor(color, ordered = FALSE)E          -95.142             
         (62.037)            
factor(color, ordered = FALSE)F          554.742 ***         
         (62.374)            
factor(color, ordered = FALSE)G          832.357 ***         
         (60.338)            
factor(color, ordered = FALSE)H          1324.183 ***         
         (64.296)            
factor(color, ordered = FALSE)I          1929.902 ***         
         (71.561)            
factor(color, ordered = FALSE)J          2164.044 ***         
         (88.144)            
N 53940         53940         53940        
R2 0.851     0.032     0.847    
logLik -472488.441     -522908.139     -26617.649    
AIC 944984.882     1045834.277     53243.298    
*** p < 0.001; ** p < 0.01; * p < 0.05.

The basic output includes estimates, standard errors and summary statistics.

Some of those variable names are hard to read. We can change them by specifying a named list of variables in the coefs argument, like this:

color_names <- paste0('factor(color, ordered = FALSE)', LETTERS[5:10])
names(color_names) <- paste('Color:', LETTERS[5:10])

huxreg(lm1, lm2, lm3, coefs = c('Carat' = 'carat', 'Depth' = 'depth', color_names))
(1) (2) (3)
Carat 7765.141 ***          1.971 ***
(14.009)             (0.004)   
Depth -102.165 *** -53.835 *** -0.018 ***
(4.635)    (11.815)    (0.001)   
Color: E          -95.142             
         (62.037)            
Color: F          554.742 ***         
         (62.374)            
Color: G          832.357 ***         
         (60.338)            
Color: H          1324.183 ***         
         (64.296)            
Color: I          1929.902 ***         
         (71.561)            
Color: J          2164.044 ***         
         (88.144)            
N 53940         53940         53940        
R2 0.851     0.032     0.847    
logLik -472488.441     -522908.139     -26617.649    
AIC 944984.882     1045834.277     53243.298    
*** p < 0.001; ** p < 0.01; * p < 0.05.

Alternatively, since the output from huxreg is just a huxtable, we could just edit its contents directly before we print it:

diamond_regs <- huxreg(lm1, lm2, lm3)
diamond_regs[seq(8, 18, 2), 1] <- paste('Color:', LETTERS[5:10])
diamond_regs
(1) (2) (3)
(Intercept) 4045.333 *** 6491.466 *** 7.313 ***
(286.205)    (730.537)    (0.074)   
carat 7765.141 ***          1.971 ***
(14.009)             (0.004)   
depth -102.165 *** -53.835 *** -0.018 ***
(4.635)    (11.815)    (0.001)   
Color: E          -95.142             
         (62.037)            
Color: F          554.742 ***         
         (62.374)            
Color: G          832.357 ***         
         (60.338)            
Color: H          1324.183 ***         
         (64.296)            
Color: I          1929.902 ***         
         (71.561)            
Color: J          2164.044 ***         
         (88.144)            
N 53940         53940         53940        
R2 0.851     0.032     0.847    
logLik -472488.441     -522908.139     -26617.649    
AIC 944984.882     1045834.277     53243.298    
*** p < 0.001; ** p < 0.01; * p < 0.05.

Of course, we aren’t limited to just changing names. We can also make our table prettier. Let’s add the “article” theme, and a vertical stripe for background colour, tweak a few details like font size, and add a caption. All of these are just standard huxtable commands.

suppressPackageStartupMessages(library(dplyr))
diamond_regs                                                         %>% 
      theme_article                                                  %>% 
      set_background_color(1:nrow(diamond_regs), evens, grey(.95)) %>% 
      set_font_size(final(), 1, 9)                                   %>% 
      set_bold(final(), 1, FALSE)                                    %>%
      set_top_border(final(), 1, 1)                                  %>%
      set_caption('Linear regressions of diamond prices')
Linear regressions of diamond prices
(1) (2) (3)
(Intercept) 4045.333 *** 6491.466 *** 7.313 ***
(286.205)    (730.537)    (0.074)   
carat 7765.141 ***          1.971 ***
(14.009)             (0.004)   
depth -102.165 *** -53.835 *** -0.018 ***
(4.635)    (11.815)    (0.001)   
Color: E          -95.142             
         (62.037)            
Color: F          554.742 ***         
         (62.374)            
Color: G          832.357 ***         
         (60.338)            
Color: H          1324.183 ***         
         (64.296)            
Color: I          1929.902 ***         
         (71.561)            
Color: J          2164.044 ***         
         (88.144)            
N 53940         53940         53940        
R2 0.851     0.032     0.847    
logLik -472488.441     -522908.139     -26617.649    
AIC 944984.882     1045834.277     53243.298    
*** p < 0.001; ** p < 0.01; * p < 0.05.

By default, standard errors are shown below coefficient estimates. To display them in a column to the right, use error_pos = 'right':

huxreg(lm1, lm3, error_pos = 'right')
(1) (2)
(Intercept) 4045.333 *** (286.205) 7.313 *** (0.074)
carat 7765.141 *** (14.009) 1.971 *** (0.004)
depth -102.165 *** (4.635) -0.018 *** (0.001)
N 53940               53940              
R2 0.851           0.847          
logLik -472488.441           -26617.649          
AIC 944984.882           53243.298          
*** p < 0.001; ** p < 0.01; * p < 0.05.

This will give column headings a column span of 2.

To display standard errors in the same cell as estimates, use error_pos = 'same':

huxreg(lm1, lm3, error_pos = 'same')
(1) (2)
(Intercept) 4045.333 *** (286.205) 7.313 *** (0.074)
carat 7765.141 *** (14.009) 1.971 *** (0.004)
depth -102.165 *** (4.635) -0.018 *** (0.001)
N 53940      53940     
R2 0.851  0.847 
logLik -472488.441  -26617.649 
AIC 944984.882  53243.298 
*** p < 0.001; ** p < 0.01; * p < 0.05.

You can change the default column headings by giving names to your models:

huxreg('Price' = lm1, 'Log price' = lm3)
Price Log price
(Intercept) 4045.333 *** 7.313 ***
(286.205)    (0.074)   
carat 7765.141 *** 1.971 ***
(14.009)    (0.004)   
depth -102.165 *** -0.018 ***
(4.635)    (0.001)   
N 53940         53940        
R2 0.851     0.847    
logLik -472488.441     -26617.649    
AIC 944984.882     53243.298    
*** p < 0.001; ** p < 0.01; * p < 0.05.

To display a particular row of summary statistics, use the statistics parameter. This should be a character vector. Valid values are anything returned from your models by broom::glance. Another valid value is "nobs", which returns the number of observations from the regression. If the statistics vector has names, these will be used for row headings:

broom::glance(lm1)
r.squared adj.r.squared sigma statistic p.value df logLik AIC BIC deviance df.residual
0.851 0.851 1.54e+03 1.54e+05 0 3 -4.72e+05 9.45e+05 9.45e+05 1.28e+11 53937
huxreg(lm1, lm3, statistics = c('# observations' = 'nobs', 'R squared' = 'r.squared', 'F statistic' = 'statistic',
  'P value' = 'p.value'))
(1) (2)
(Intercept) 4045.333 *** 7.313 ***
(286.205)    (0.074)   
carat 7765.141 *** 1.971 ***
(14.009)    (0.004)   
depth -102.165 *** -0.018 ***
(4.635)    (0.001)   
# observations 53940         53940        
R squared 0.851     0.847    
F statistic 153634.765     149771.327    
P value 0.000     0.000    
*** p < 0.001; ** p < 0.01; * p < 0.05.

By default, huxreg displays significance stars. You can alter the symbols used and significance levels with the stars parameter, or set stars = NULL to turn off significance stars completely.

huxreg(lm1, lm3, stars = c(`*` = 0.1, `**` = 0.05, `***` = 0.01)) # a little boastful?
(1) (2)
(Intercept) 4045.333 *** 7.313 ***
(286.205)    (0.074)   
carat 7765.141 *** 1.971 ***
(14.009)    (0.004)   
depth -102.165 *** -0.018 ***
(4.635)    (0.001)   
N 53940         53940        
R2 0.851     0.847    
logLik -472488.441     -26617.649    
AIC 944984.882     53243.298    
*** p < 0.01; ** p < 0.05; * p < 0.1.
huxreg(lm1, lm3, stars = NULL) 
(1) (2)
(Intercept) 4045.333  7.313 
(286.205) (0.074)
carat 7765.141  1.971 
(14.009) (0.004)
depth -102.165  -0.018 
(4.635) (0.001)
N 53940      53940     
R2 0.851  0.847 
logLik -472488.441  -26617.649 
AIC 944984.882  53243.298 
.

You aren’t limited to displaying standard errors of the estimates. If you prefer, you can display t statistics or p values, using the error_format option. Any column from tidy can be used by putting it in curly brackets:

huxreg(lm1, lm3, error_format = '({statistic})')
(1) (2)
(Intercept) 4045.333 *** 7.313 ***
(14.134)    (99.383)   
carat 7765.141 *** 1.971 ***
(554.282)    (547.305)   
depth -102.165 *** -0.018 ***
(-22.041)    (-14.936)   
N 53940         53940        
R2 0.851     0.847    
logLik -472488.441     -26617.649    
AIC 944984.882     53243.298    
*** p < 0.001; ** p < 0.01; * p < 0.05.
huxreg(lm1, lm3, error_format = '({p.value})')
(1) (2)
(Intercept) 4045.333 *** 7.313 ***
(0.000)    (0.000)   
carat 7765.141 *** 1.971 ***
(0.000)    (0.000)   
depth -102.165 *** -0.018 ***
(0.000)    (0.000)   
N 53940         53940        
R2 0.851     0.847    
logLik -472488.441     -26617.649    
AIC 944984.882     53243.298    
*** p < 0.001; ** p < 0.01; * p < 0.05.

Or you can display confidence intervals. Use ci_level to set the confidence level for the interval, then use {conf.low} and {conf.high} in error_format:

huxreg(lm1, lm3, ci_level = .99, error_format = '{conf.low} to {conf.high}')
(1) (2)
(Intercept) 4045.333 *** 7.313 ***
3308.091 to 4782.576     7.123 to 7.502    
carat 7765.141 *** 1.971 ***
7729.054 to 7801.228     1.962 to 1.981    
depth -102.165 *** -0.018 ***
-114.105 to -90.225     -0.021 to -0.015    
N 53940         53940        
R2 0.851     0.847    
logLik -472488.441     -26617.649    
AIC 944984.882     53243.298    
*** p < 0.001; ** p < 0.01; * p < 0.05.

To change the footnote, use note. If note contains the string "{stars}" it will be replaced by a description of the significance stars used. If you don’t want a footnote, just set note = NULL.

huxreg(lm1, lm3, note = 'Linear regressions on diamond price. {stars}.')
(1) (2)
(Intercept) 4045.333 *** 7.313 ***
(286.205)    (0.074)   
carat 7765.141 *** 1.971 ***
(14.009)    (0.004)   
depth -102.165 *** -0.018 ***
(4.635)    (0.001)   
N 53940         53940        
R2 0.851     0.847    
logLik -472488.441     -26617.649    
AIC 944984.882     53243.298    
Linear regressions on diamond price. *** p < 0.001; ** p < 0.01; * p < 0.05.

To change number formatting, set the number_format parameter. This works the same as the number_format property for a huxtable - if it is numeric, numbers will be rounded to that many decimal places; if it is character, it will be taken as a format to the base R sprintf function. huxreg tries to be smart and to format summary statistics like nobs as integers.

huxreg(lm1, lm3, number_format = 2)
(1) (2)
(Intercept) 4045.33 *** 7.31 ***
(286.21)    (0.07)   
carat 7765.14 *** 1.97 ***
(14.01)    (0.00)   
depth -102.17 *** -0.02 ***
(4.64)    (0.00)   
N 53940        53940       
R2 0.85     0.85    
logLik -472488.44     -26617.65    
AIC 944984.88     53243.30    
*** p < 0.001; ** p < 0.01; * p < 0.05.

Lastly, if you want to bold all significant coefficients, set the parameter bold_signif to a maximum significance level:

huxreg(lm1, lm3, bold_signif = 0.05)
(1) (2)
(Intercept) 4045.333 *** 7.313 ***
(286.205)    (0.074)   
carat 7765.141 *** 1.971 ***
(14.009)    (0.004)   
depth -102.165 *** -0.018 ***
(4.635)    (0.001)   
N 53940         53940        
R2 0.851     0.847    
logLik -472488.441     -26617.649    
AIC 944984.882     53243.298    
*** p < 0.001; ** p < 0.01; * p < 0.05.

Altering data

Sometimes, you want to report different statistics for a model. For example, you might want to use robust standard errors.

One way to do this is to pass a tidy-able test object into huxreg. The function coeftest in the “lmtest” package has tidy methods defined:

library(lmtest)
## Loading required package: zoo
## 
## Attaching package: 'zoo'
## The following objects are masked from 'package:base':
## 
##     as.Date, as.Date.numeric
library(sandwich)
lm_robust <- coeftest(lm1, vcov = vcovHC)
huxreg("Normal SEs" = lm1, "Robust SEs" = lm_robust)
## Warning in FUN(X[[i]], ...): No `glance` method for model of class coeftest
Normal SEs Robust SEs
(Intercept) 4045.333 *** 4045.333 ***
(286.205)    (369.327)   
carat 7765.141 *** 7765.141 ***
(14.009)    (25.114)   
depth -102.165 *** -102.165 ***
(4.635)    (5.948)   
N 53940                 
R2 0.851             
logLik -472488.441             
AIC 944984.882             
*** p < 0.001; ** p < 0.01; * p < 0.05.

If that is not possible, you can compute statistics yourself and add them to your model using the tidy_override function:

lm_fixed <- tidy_override(lm1, p.value = c(0.5, 0.2, 0.06))
huxreg("Normal p values" = lm1, "Supplied p values" = lm_fixed)
Normal p values Supplied p values
(Intercept) 4045.333 *** 4045.333 
(286.205)    (286.205)
carat 7765.141 *** 7765.141 
(14.009)    (14.009)
depth -102.165 *** -102.165 
(4.635)    (4.635)
N 53940         53940     
R2 0.851     0.851 
logLik -472488.441     -472488.441 
AIC 944984.882     944984.882 
*** p < 0.001; ** p < 0.01; * p < 0.05.

You can override any statistics returned by tidy or glance.