mtable and stargazer

mtable_stargazer.R

library(memisc)
library(stargazer)
library(pander)

lm0 <- lm(sr ~ pop15 + pop75, data = LifeCycleSavings)
lm1 <- lm(sr ~ dpi + ddpi, data = LifeCycleSavings)
lm2 <- lm(sr ~ pop15 + pop75 + dpi + ddpi, data = LifeCycleSavings)

mtable123 <- mtable("Model 1"=lm0,"Model 2"=lm1,"Model 3"=lm2, summary.stats=c("sigma","R-squared","F","p","N"))
pander(mtable123)

# -----------------------------------------------------
#      &nbsp;         Model 1     Model 2     Model 3  
# ----------------- ----------- ----------- -----------
#  **(Intercept)**  30.628***\  6.360***\   28.566***\ 
#                     (7.409)    (1.252)      (7.355)  

#     **pop15**     -0.471**\       \       -0.461**\  
#                    (0.147)                 (0.145)   

#     **pop75**      -1.934\        \        -1.691\   
#                   (1.041)                 (1.084)    

#      **dpi**          \         0.001\     -0.000\   
#                               (0.001)     (0.001)    

#     **ddpi**          \        0.529*\     0.410*\   
#                               (0.210)     (0.196)    

#     **sigma**        3.931       4.189       3.803   

#   **R-squared**      0.262       0.162       0.338   

#       **F**          8.332       4.528       5.756   

#       **p**          0.001       0.016       0.001   

#       **N**           50          50          50     
# -----------------------------------------------------


stargazer(lm0, lm1, lm2, type="text", 
          star.cutoffs=c(0.05, 0.01, 0.001), model.names=FALSE, 
          dep.var.labels.include=FALSE, dep.var.caption="", 
          notes=c("Here are some useful notes about the results", 
                  "Like stuff about robust standard errors and whatnot"))

# ===================================================================================
#                              (1)                  (2)                  (3)         
# -----------------------------------------------------------------------------------
# pop15                     -0.471**                                  -0.461**       
#                            (0.147)                                   (0.145)       
                                                                                   
# pop75                      -1.934                                    -1.691        
#                            (1.041)                                   (1.084)       
                                                                                   
# dpi                                              0.001               -0.0003       
#                                                 (0.001)              (0.001)       
                                                                                   
# ddpi                                            0.529*               0.410*        
#                                                 (0.210)              (0.196)       
                                                                                   
# Constant                  30.628***            6.360***             28.566***      
#                            (7.409)              (1.252)              (7.355)       
                                                                                   
# -----------------------------------------------------------------------------------
# Observations                 50                   50                   50          
# R2                          0.262                0.162                0.338        
# Adjusted R2                 0.230                0.126                0.280        
# Residual Std. Error    3.931 (df = 47)      4.189 (df = 47)      3.803 (df = 45)   
# F Statistic         8.332*** (df = 2; 47) 4.528* (df = 2; 47) 5.756*** (df = 4; 45)
# ===================================================================================
# Note:                                                 *p<0.05; **p<0.01; ***p<0.001
#                                        Here are some useful notes about the results
#                                 Like stuff about robust standard errors and whatnot

The problem is that Pandoc's markdown has a limited support for tables -- so that it can be truly compatible with most document formats. E.g. it does not support col/row spanning and borders, that's why Roman and I came up with the workaround of putting two values in a cell with the help of so called "hard breaks" inside of cells. Notes below tables are not supported similarly, only captions can be added.

If we want the Observations/R2 etc to be separated from the other model parameters, it could be returned in a separate table, like it's done for a simple linear model:

> pander(summary(lm0))

--------------------------------------------------------------
              Estimate   Std. Error   t value   Pr(>|t|) 
----------------- ---------- ------------ --------- ----------
    **pop15**      -0.4708      0.1468     -3.207    0.002414 

    **pop75**       -1.934      1.041      -1.858    0.06943  

 **(Intercept)**    30.63       7.409       4.134   0.0001457 
--------------------------------------------------------------


-------------------------------------------------------------
 Observations   Residual Std. Error   $R^2$   Adjusted $R^2$ 
-------------- --------------------- ------- ----------------
      50               3.931         0.2617       0.2303     
-------------------------------------------------------------

Table: Fitting linear model: sr ~ pop15 + pop75

And it comes with some minor functions as well:

> pander(summary(lm0), covariate.labels = c('foo', 'bar'), omit = '(Intercept)', summary = FALSE, caption = 'foobar!', add.significance.stars = TRUE)

----------------------------------------------------------
      Estimate   Std. Error   t value     Pr(>|t|)   
--------- ---------- ------------ --------- --------------
 **foo**   -0.4708      0.1468     -3.207   _0.002414_ * *

 **bar**    -1.934      1.041      -1.858     _0.06943_   
----------------------------------------------------------

Table: foobar!

It seems to me that if you would like to get stargazer-like markdown output, then it requires same trade-off (as seen above), and it has to be implemented separately from stargazer as it returns ASCII that cannot be (easily) post-processed. But please let me know what features you miss, and I will be happy to try to resolve those.