Skip to contents

Directly obtain the regression results

Source: R/regress.R
regress.Rd

These functions are wrappers for the broom package to directly obtain the regression results. The naming convention is regress_* for raw formula and tidy_* for the model object.

  • regress() returns the regression results as a list.

  • regress_coef() returns the coefficients.

  • regress_stat() returns the model statistics.

  • regress_data() returns the augmented data.

Usage

regress(.data, formula, model = "lm", digit = integer(0), format = "%.3f")

regress_coef(
  .data,
  formula,
  model = "lm",
  digit = integer(0),
  format = "%.3f"
)

regress_stat(.data, formula, model = "lm")

regress_data(.data, formula, model = "lm")

Arguments

.data

A data frame, data frame extension (e.g. a tibble), or a lazy data frame (e.g. from dbplyr or dtplyr).

formula

A formula specifying the model.

model

A character string specifying the model to be used.

digit

A positive integer specifying the number of digits to be displayed.

format

A character string specifying the format of the coefficients. The default is %.3f, which means three decimal places.

Value

Return the regression results as a list or a tibble.

See also

tidy_coef()

Examples

regress(mtcars, mpg ~ wt + hp)
#> $coef
#> # A tibble: 3 × 7
#>   term        coef                 estimate std.error statistic  p.value star 
#>   <chr>       <chr>                   <dbl>     <dbl>     <dbl>    <dbl> <chr>
#> 1 (Intercept) "37.227***\n(1.599)"  37.2      1.60        23.3  2.57e-20 ***  
#> 2 wt          "-3.878***\n(0.633)"  -3.88     0.633       -6.13 1.12e- 6 ***  
#> 3 hp          "-0.032**\n(0.009)"   -0.0318   0.00903     -3.52 1.45e- 3 **   
#> 
#> $stat
#> # A tibble: 1 × 12
#>   r.squared adj.r.squared sigma statistic  p.value    df logLik   AIC   BIC
#>       <dbl>         <dbl> <dbl>     <dbl>    <dbl> <dbl>  <dbl> <dbl> <dbl>
#> 1     0.827         0.815  2.59      69.2 9.11e-12     2  -74.3  157.  163.
#> # ℹ 3 more variables: deviance <dbl>, df.residual <int>, nobs <int>
#> 
#> $data
#> # A tibble: 32 × 10
#>    .rownames     mpg    wt    hp .fitted .resid   .hat .sigma .cooksd .std.resid
#>    <chr>       <dbl> <dbl> <dbl>   <dbl>  <dbl>  <dbl>  <dbl>   <dbl>      <dbl>
#>  1 Mazda RX4    21    2.62   110    23.6 -2.57  0.0443   2.59 1.59e-2    -1.01  
#>  2 Mazda RX4 …  21    2.88   110    22.6 -1.58  0.0405   2.62 5.46e-3    -0.623 
#>  3 Datsun 710   22.8  2.32    93    25.3 -2.48  0.0602   2.59 2.07e-2    -0.985 
#>  4 Hornet 4 D…  21.4  3.22   110    21.3  0.135 0.0475   2.64 4.72e-5     0.0533
#>  5 Hornet Spo…  18.7  3.44   175    18.3  0.373 0.0369   2.64 2.74e-4     0.146 
#>  6 Valiant      18.1  3.46   105    20.5 -2.37  0.0672   2.60 2.16e-2    -0.948 
#>  7 Duster 360   14.3  3.57   245    15.6 -1.30  0.117    2.63 1.26e-2    -0.533 
#>  8 Merc 240D    24.4  3.19    62    22.9  1.51  0.116    2.62 1.68e-2     0.620 
#>  9 Merc 230     22.8  3.15    95    22.0  0.806 0.0600   2.63 2.19e-3     0.321 
#> 10 Merc 280     19.2  3.44   123    20.0 -0.779 0.0469   2.63 1.55e-3    -0.308 
#> # ℹ 22 more rows
#> 
regress_coef(mtcars, mpg ~ wt + hp)
#> # A tibble: 3 × 7
#>   term        coef                 estimate std.error statistic  p.value star 
#>   <chr>       <chr>                   <dbl>     <dbl>     <dbl>    <dbl> <chr>
#> 1 (Intercept) "37.227***\n(1.599)"  37.2      1.60        23.3  2.57e-20 ***  
#> 2 wt          "-3.878***\n(0.633)"  -3.88     0.633       -6.13 1.12e- 6 ***  
#> 3 hp          "-0.032**\n(0.009)"   -0.0318   0.00903     -3.52 1.45e- 3 **   

mtcars %>%
  dplyr::mutate(mpg20 = mpg > 20) %>%
  regress(mpg20 ~ wt + hp, model = "logit")
#> Warning: glm.fit: algorithm did not converge
#> Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
#> Warning: glm.fit: algorithm did not converge
#> Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
#> Warning: glm.fit: algorithm did not converge
#> Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
#> $coef
#> # A tibble: 3 × 7
#>   term        coef                    estimate std.error statistic p.value star 
#>   <chr>       <chr>                      <dbl>     <dbl>     <dbl>   <dbl> <chr>
#> 1 (Intercept) "894.228\n(365884.162)"   894.     365884.   0.00244   0.998 ""   
#> 2 wt          "-202.865\n(84688.218)"  -203.      84688.  -0.00240   0.998 ""   
#> 3 hp          "-2.021\n(858.062)"        -2.02      858.  -0.00236   0.998 ""   
#> 
#> $stat
#> # A tibble: 1 × 8
#>   null.deviance df.null   logLik   AIC   BIC     deviance df.residual  nobs
#>           <dbl>   <int>    <dbl> <dbl> <dbl>        <dbl>       <int> <int>
#> 1          43.9      31 -5.58e-9  6.00  10.4 0.0000000112          29    32
#> 
#> $data
#> # A tibble: 32 × 10
#>    .rownames       mpg20    wt    hp .fitted   .resid    .hat    .sigma  .cooksd
#>    <chr>           <lgl> <dbl> <dbl>   <dbl>    <dbl>   <dbl>     <dbl>    <dbl>
#>  1 Mazda RX4       TRUE   2.62   110   140.   2.11e-8 7.71e-7   2.00e-5 5.70e-23
#>  2 Mazda RX4 Wag   TRUE   2.88   110    88.7  2.11e-8 3.15e-7   2.00e-5 2.33e-23
#>  3 Datsun 710      TRUE   2.32    93   236.   2.11e-8 2.10e-6   2.00e-5 1.55e-22
#>  4 Hornet 4 Drive  TRUE   3.22   110    19.7  7.48e-5 1.00e+0 NaN       3.12e- 1
#>  5 Hornet Sportab… FALSE  3.44   175  -157.  -2.11e-8 9.55e-7   2.00e-5 7.07e-23
#>  6 Valiant         FALSE  3.46   105   -19.9 -6.74e-5 1.00e+0 NaN       3.53e- 1
#>  7 Duster 360      FALSE  3.57   245  -325.  -2.11e-8 4.06e-6   2.00e-5 3.00e-22
#>  8 Merc 240D       TRUE   3.19    62   122.   2.11e-8 5.89e-7   2.00e-5 4.36e-23
#>  9 Merc 230        TRUE   3.15    95    63.2  2.11e-8 1.64e-7   2.00e-5 1.21e-23
#> 10 Merc 280        FALSE  3.44   123   -52.2 -2.11e-8 1.15e-7   2.00e-5 8.48e-24
#> # ℹ 22 more rows
#> # ℹ 1 more variable: .std.resid <dbl>
#>