Summary information for fixest_multi objects. In particular, this is used to specify the type of standard-errors to be computed.
# S3 method for class 'fixest_multi'
summary(
object,
type = "short",
vcov = NULL,
se = NULL,
cluster = NULL,
ssc = NULL,
.vcov = NULL,
stage = 2,
lean = FALSE,
n = 1000,
...
)A fixest_multi object, obtained from a fixest estimation leading to
multiple results.
A character either equal to "short", "long", "compact", "se_compact"
or "se_long". If short, only the table of coefficients is displayed for each estimation.
If long, then the full results are displayed for each estimation. If compact,
a data.frame is returned with one line per model and the formatted
coefficients + standard-errors in the columns. If se_compact, a data.frame is
returned with one line per model, one numeric column for each coefficient and one numeric
column for each standard-error. If "se_long", same as "se_compact" but the data is in a
long format instead of wide.
Versatile argument to specify the VCOV. In general, it is either a character
scalar equal to a VCOV type, either a formula of the form: vcov_type ~ variables. The
VCOV types implemented are: "iid", "hetero" (or "HC1"), "cluster", "twoway",
"NW" (or "newey_west"), "DK" (or "driscoll_kraay"), and "conley". It also accepts
object from vcov_cluster, vcov_NW, NW,
vcov_DK, DK, vcov_conley and
conley. It also accepts covariance matrices computed externally.
Finally it accepts functions to compute the covariances. See the vcov documentation
in the vignette.
Character scalar. Which kind of standard error should be computed:
“standard”, “hetero”, “cluster”, “twoway”, “threeway”
or “fourway”? By default if there are clusters in the estimation:
se = "cluster", otherwise se = "iid". Note that this argument is deprecated,
you should use vcov instead.
Tells how to cluster the standard-errors (if clustering is requested).
Can be either a list of vectors, a character vector of variable names, a formula or
an integer vector. Assume we want to perform 2-way clustering over var1 and var2
contained in the data.frame base used for the estimation. All the following
cluster arguments are valid and do the same thing:
cluster = base[, c("var1", "var2")], cluster = c("var1", "var2"), cluster = ~var1+var2.
If the two variables were used as fixed-effects in the estimation, you can leave it
blank with vcov = "twoway" (assuming var1 [resp. var2] was
the 1st [resp. 2nd] fixed-effect). You can interact two variables using ^ with
the following syntax: cluster = ~var1^var2 or cluster = "var1^var2".
An object of class ssc.type obtained with the function ssc. Represents
how the degree of freedom correction should be done.You must use the function ssc
for this argument. The arguments and defaults of the function ssc are:
adj = TRUE, fixef.K="nested", cluster.adj = TRUE, cluster.df = "min",
t.df = "min", fixef.force_exact=FALSE). See the help of the function ssc for details.
Can be equal to 2 (default), 1, 1:2 or 2:1. Only used if the object
is an IV estimation: defines the stage to which summary should be applied. If stage = 1
and there are multiple endogenous regressors or if stage is of length 2, then an
object of class fixest_multi is returned.
Logical, default is FALSE. Used to reduce the (memory) size of the summary object.
If TRUE, then all objects of length N (the number of observations) are removed
from the result. Note that some fixest methods may consequently not work when applied
to the summary.
Integer, default is 1000. Number of coefficients to display when the print method is used.
Not currently used.
It returns either an object of class fixest_multi (if type equals short or long),
either a data.frame (if type equals compact or se_compact).
The main fixest estimation functions: feols, fepois,
fenegbin, feglm, feNmlm. Tools for mutliple fixest
estimations: summary.fixest_multi, print.fixest_multi, as.list.fixest_multi,
sub-sub-.fixest_multi, sub-.fixest_multi.
base = iris
names(base) = c("y", "x1", "x2", "x3", "species")
# Multiple estimation
res = feols(y ~ csw(x1, x2, x3), base, split = ~species)
# By default, the type is "short"
# You can still use the arguments from summary.fixest
summary(res, se = "hetero")
#> Standard-errors: Heteroskedasticity-robust
#>
#> ### Sample: setosa
#>
#> Expl. vars.: x1
#> Estimate Std. Error t value Pr(>|t|)
#> (Intercept) 2.639001 0.298624 8.83722 1.2326e-11 ***
#> x1 0.690490 0.085903 8.03803 1.9293e-10 ***
#> ---
#> Expl. vars.: x1 + x2
#> Estimate Std. Error t value Pr(>|t|)
#> (Intercept) 2.303738 0.433561 5.31352 2.8928e-06 ***
#> x1 0.667416 0.092247 7.23508 3.6001e-09 ***
#> x2 0.283419 0.264794 1.07034 2.8993e-01
#> ---
#> Expl. vars.: x1 + x2 + x3
#> Estimate Std. Error t value Pr(>|t|)
#> (Intercept) 2.351890 0.437230 5.379067 2.4390e-06 ***
#> x1 0.654835 0.092468 7.081774 6.8711e-09 ***
#> x2 0.237560 0.275270 0.863009 3.9261e-01
#> x3 0.252126 0.284622 0.885827 3.8032e-01
#>
#> ### Sample: versicolor
#>
#> Expl. vars.: x1
#> Estimate Std. Error t value Pr(>|t|)
#> (Intercept) 3.539735 0.617019 5.73683 6.3108e-07 ***
#> x1 0.865078 0.220701 3.91969 2.8079e-04 ***
#> ---
#> Expl. vars.: x1 + x2
#> Estimate Std. Error t value Pr(>|t|)
#> (Intercept) 2.116431 0.431296 4.90714 1.1507e-05 ***
#> x1 0.247642 0.176500 1.40307 1.6717e-01
#> x2 0.735587 0.111386 6.60395 3.2636e-08 ***
#> ---
#> Expl. vars.: x1 + x2 + x3
#> Estimate Std. Error t value Pr(>|t|)
#> (Intercept) 1.895540 0.434294 4.36465 7.1434e-05 ***
#> x1 0.386858 0.207117 1.86782 6.8165e-02 .
#> x2 0.908337 0.159800 5.68420 8.5897e-07 ***
#> x3 -0.679224 0.436600 -1.55571 1.2663e-01
#>
#> ### Sample: virginica
#>
#> Expl. vars.: x1
#> Estimate Std. Error t value Pr(>|t|)
#> (Intercept) 3.906836 0.760748 5.13552 5.0735e-06 ***
#> x1 0.901534 0.246200 3.66179 6.2338e-04 ***
#> ---
#> Expl. vars.: x1 + x2
#> Estimate Std. Error t value Pr(>|t|)
#> (Intercept) 0.624782 0.538060 1.16118 2.5143e-01
#> x1 0.259954 0.130150 1.99734 5.1596e-02 .
#> x2 0.934819 0.076962 12.14650 4.2031e-16 ***
#> ---
#> Expl. vars.: x1 + x2 + x3
#> Estimate Std. Error t value Pr(>|t|)
#> (Intercept) 0.699883 0.553402 1.264692 2.1235e-01
#> x1 0.330337 0.122494 2.696759 9.7515e-03 **
#> x2 0.945536 0.080881 11.690520 2.2562e-15 ***
#> x3 -0.169753 0.210310 -0.807154 4.2373e-01
summary(res, type = "long")
#>
#> ### Sample: setosa
#>
#> Expl. vars.: x1
#> OLS estimation, Dep. Var.: y
#> Observations: 50
#> Sample (species): setosa
#> Standard-errors: IID
#> Estimate Std. Error t value Pr(>|t|)
#> (Intercept) 2.639001 0.310014 8.51251 3.7424e-11 ***
#> x1 0.690490 0.089899 7.68074 6.7098e-10 ***
#> ---
#> Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
#> RMSE: 0.233723 Adj. R2: 0.542029
#>
#> Expl. vars.: x1 + x2
#> OLS estimation, Dep. Var.: y
#> Observations: 50
#> Sample (species): setosa
#> Standard-errors: IID
#> Estimate Std. Error t value Pr(>|t|)
#> (Intercept) 2.303738 0.385294 5.97917 2.8943e-07 ***
#> x1 0.667416 0.090356 7.38653 2.1252e-09 ***
#> x2 0.283419 0.197224 1.43704 1.5733e-01
#> ---
#> Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
#> RMSE: 0.228751 Adj. R2: 0.551971
#>
#> Expl. vars.: x1 + x2 + x3
#> OLS estimation, Dep. Var.: y
#> Observations: 50
#> Sample (species): setosa
#> Standard-errors: IID
#> Estimate Std. Error t value Pr(>|t|)
#> (Intercept) 2.351890 0.392868 5.986471 3.0342e-07 ***
#> x1 0.654835 0.092447 7.083324 6.8344e-09 ***
#> x2 0.237560 0.208019 1.142011 2.5936e-01
#> x3 0.252126 0.346864 0.726873 4.7099e-01
#> ---
#> Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
#> RMSE: 0.227449 Adj. R2: 0.547429
#>
#> ### Sample: versicolor
#>
#> Expl. vars.: x1
#> OLS estimation, Dep. Var.: y
#> Observations: 50
#> Sample (species): versicolor
#> Standard-errors: IID
#> Estimate Std. Error t value Pr(>|t|)
#> (Intercept) 3.539735 0.562874 6.28869 9.0690e-08 ***
#> x1 0.865078 0.201938 4.28389 8.7719e-05 ***
#> ---
#> Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
#> RMSE: 0.434612 Adj. R2: 0.261511
#>
#> Expl. vars.: x1 + x2
#> OLS estimation, Dep. Var.: y
#> Observations: 50
#> Sample (species): versicolor
#> Standard-errors: IID
#> Estimate Std. Error t value Pr(>|t|)
#> (Intercept) 2.116431 0.494256 4.28206 9.0640e-05 ***
#> x1 0.247642 0.186839 1.32543 1.9144e-01
#> x2 0.735587 0.124768 5.89565 3.8707e-07 ***
#> ---
#> Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
#> RMSE: 0.329521 Adj. R2: 0.566438
#>
#> Expl. vars.: x1 + x2 + x3
#> OLS estimation, Dep. Var.: y
#> Observations: 50
#> Sample (species): versicolor
#> Standard-errors: IID
#> Estimate Std. Error t value Pr(>|t|)
#> (Intercept) 1.895540 0.507055 3.73833 5.1122e-04 ***
#> x1 0.386858 0.204545 1.89131 6.4890e-02 .
#> x2 0.908337 0.165432 5.49068 1.6667e-06 ***
#> x3 -0.679224 0.435382 -1.56006 1.2560e-01
#> ---
#> Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
#> RMSE: 0.321135 Adj. R2: 0.579273
#>
#> ### Sample: virginica
#>
#> Expl. vars.: x1
#> OLS estimation, Dep. Var.: y
#> Observations: 50
#> Sample (species): virginica
#> Standard-errors: IID
#> Estimate Std. Error t value Pr(>|t|)
#> (Intercept) 3.906836 0.757061 5.16053 4.6563e-06 ***
#> x1 0.901534 0.253106 3.56189 8.4346e-04 ***
#> ---
#> Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
#> RMSE: 0.559836 Adj. R2: 0.192579
#>
#> Expl. vars.: x1 + x2
#> OLS estimation, Dep. Var.: y
#> Observations: 50
#> Sample (species): virginica
#> Standard-errors: IID
#> Estimate Std. Error t value Pr(>|t|)
#> (Intercept) 0.624782 0.524867 1.19036 2.3988e-01
#> x1 0.259954 0.153338 1.69531 9.6634e-02 .
#> x2 0.934819 0.089602 10.43302 8.0094e-14 ***
#> ---
#> Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
#> RMSE: 0.307439 Adj. R2: 0.75132
#>
#> Expl. vars.: x1 + x2 + x3
#> OLS estimation, Dep. Var.: y
#> Observations: 50
#> Sample (species): virginica
#> Standard-errors: IID
#> Estimate Std. Error t value Pr(>|t|)
#> (Intercept) 0.699883 0.533601 1.311623 1.9616e-01
#> x1 0.330337 0.174329 1.894909 6.4400e-02 .
#> x2 0.945536 0.090722 10.422336 1.0743e-13 ***
#> x3 -0.169753 0.198072 -0.857023 3.9587e-01
#> ---
#> Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
#> RMSE: 0.305014 Adj. R2: 0.749908
summary(res, type = "compact")
#> lhs sample rhs (Intercept) x1 x2
#> 1 y setosa x1 2.64*** (0.310) 0.690*** (0.090)
#> 2 y setosa x1 + x2 2.30*** (0.385) 0.667*** (0.090) 0.283 (0.197)
#> 3 y setosa x1 + x2 + x3 2.35*** (0.393) 0.655*** (0.092) 0.238 (0.208)
#> 4 y versicolor x1 3.54*** (0.563) 0.865*** (0.202)
#> 5 y versicolor x1 + x2 2.12*** (0.494) 0.248 (0.187) 0.736*** (0.125)
#> 6 y versicolor x1 + x2 + x3 1.90*** (0.507) 0.387. (0.205) 0.908*** (0.165)
#> 7 y virginica x1 3.91*** (0.757) 0.902*** (0.253)
#> 8 y virginica x1 + x2 0.625 (0.525) 0.260. (0.153) 0.935*** (0.090)
#> 9 y virginica x1 + x2 + x3 0.700 (0.534) 0.330. (0.174) 0.946*** (0.091)
#> x3
#> 1
#> 2
#> 3 0.252 (0.347)
#> 4
#> 5
#> 6 -0.679 (0.435)
#> 7
#> 8
#> 9 -0.170 (0.198)
summary(res, type = "se_compact")
#> lhs sample rhs (Intercept) (Intercept)__se x1 x1__se
#> 1 y setosa x1 2.6390012 0.3100143 0.6904897 0.08989888
#> 2 y setosa x1 + x2 2.3037382 0.3852942 0.6674162 0.09035581
#> 3 y setosa x1 + x2 + x3 2.3518898 0.3928675 0.6548350 0.09244742
#> 4 y versicolor x1 3.5397347 0.5628736 0.8650777 0.20193757
#> 5 y versicolor x1 + x2 2.1164314 0.4942556 0.2476422 0.18683892
#> 6 y versicolor x1 + x2 + x3 1.8955395 0.5070552 0.3868576 0.20454490
#> 7 y virginica x1 3.9068365 0.7570605 0.9015345 0.25310551
#> 8 y virginica x1 + x2 0.6247824 0.5248675 0.2599540 0.15333757
#> 9 y virginica x1 + x2 + x3 0.6998830 0.5336009 0.3303370 0.17432873
#> x2 x2__se x3 x3__se
#> 1 NA NA NA NA
#> 2 0.2834193 0.19722377 NA NA
#> 3 0.2375602 0.20801921 0.2521257 0.3468636
#> 4 NA NA NA NA
#> 5 0.7355868 0.12476776 NA NA
#> 6 0.9083370 0.16543248 -0.6792238 0.4353821
#> 7 NA NA NA NA
#> 8 0.9348189 0.08960197 NA NA
#> 9 0.9455356 0.09072204 -0.1697527 0.1980724
summary(res, type = "se_long")
#> lhs sample rhs type (Intercept) x1 x2
#> 1 y setosa x1 coef 2.6390012 0.69048972 NA
#> 2 y setosa x1 se 0.3100143 0.08989888 NA
#> 3 y setosa x1 + x2 coef 2.3037382 0.66741621 0.28341929
#> 4 y setosa x1 + x2 se 0.3852942 0.09035581 0.19722377
#> 5 y setosa x1 + x2 + x3 coef 2.3518898 0.65483497 0.23756017
#> 6 y setosa x1 + x2 + x3 se 0.3928675 0.09244742 0.20801921
#> 7 y versicolor x1 coef 3.5397347 0.86507772 NA
#> 8 y versicolor x1 se 0.5628736 0.20193757 NA
#> 9 y versicolor x1 + x2 coef 2.1164314 0.24764216 0.73558681
#> 10 y versicolor x1 + x2 se 0.4942556 0.18683892 0.12476776
#> 11 y versicolor x1 + x2 + x3 coef 1.8955395 0.38685762 0.90833700
#> 12 y versicolor x1 + x2 + x3 se 0.5070552 0.20454490 0.16543248
#> 13 y virginica x1 coef 3.9068365 0.90153448 NA
#> 14 y virginica x1 se 0.7570605 0.25310551 NA
#> 15 y virginica x1 + x2 coef 0.6247824 0.25995398 0.93481889
#> 16 y virginica x1 + x2 se 0.5248675 0.15333757 0.08960197
#> 17 y virginica x1 + x2 + x3 coef 0.6998830 0.33033703 0.94553559
#> 18 y virginica x1 + x2 + x3 se 0.5336009 0.17432873 0.09072204
#> x3
#> 1 NA
#> 2 NA
#> 3 NA
#> 4 NA
#> 5 0.2521257
#> 6 0.3468636
#> 7 NA
#> 8 NA
#> 9 NA
#> 10 NA
#> 11 -0.6792238
#> 12 0.4353821
#> 13 NA
#> 14 NA
#> 15 NA
#> 16 NA
#> 17 -0.1697527
#> 18 0.1980724