Hi,
I want to compare two nested models in stargaze. There is a way to do so by using keep.stats.
Example:
# Set the seed for reproducibility
set.seed(123)
# Generate data
Modelling_fw_data <- data.frame(
x1 = runif(100, min = 1, max = 10),
x2 = runif(100, min = 1, max = 10),
x3 = runif(100, min = 1, max = 10),
x4 = runif(100, min = 1, max = 10),
y = runif(100, min = 1, max = 10)
)
GLM_modelling_fw_example1 <- glm(
y ~ log(x1)
,data = Modelling_fw_data ,family = inverse.gaussian(link = "log") ,weights = x4)
GLM_modelling_fw_example2 <- glm(
y ~ log(x1) + x2 + x3 ,data = Modelling_fw_data ,family = inverse.gaussian(link = "log") ,weights = x4)
stargazer(GLM_modelling_fw_example1 ,GLM_modelling_fw_example2
,keep.stat = c('all')
,title='CM & RM model summary'
,model.names = FALSE
,column.labels=c('CM', 'RM')
,align=TRUE
,type = "html")
The above stargazer prints non of the statistics. Does anyone know why is that?
Thanks
I'm not sure, but here's an alternative
library(broom)
set.seed(123)
Modelling_fw_data <- data.frame(
x1 = runif(100, min = 1, max = 10),
x2 = runif(100, min = 1, max = 10),
x3 = runif(100, min = 1, max = 10),
x4 = runif(100, min = 1, max = 10),
y = runif(100, min = 1, max = 10)
)
example1 <- glm(
y ~ log(x1),
data = Modelling_fw_data, family = inverse.gaussian(link = "log"), weights = x4
)
example2 <- glm(
y ~ log(x1) + x2 + x3,
data = Modelling_fw_data, family = inverse.gaussian(link = "log"), weights = x4
)
tidy(example1)
#> # A tibble: 2 × 5
#> term estimate std.error statistic p.value
#> <chr> <dbl> <dbl> <dbl> <dbl>
#> 1 (Intercept) 2.00 0.150 13.3 1.07e-23
#> 2 log(x1) -0.182 0.0889 -2.05 4.33e- 2
tidy(example2)
#> # A tibble: 4 × 5
#> term estimate std.error statistic p.value
#> <chr> <dbl> <dbl> <dbl> <dbl>
#> 1 (Intercept) 2.09 0.239 8.74 7.68e-14
#> 2 log(x1) -0.204 0.0915 -2.23 2.84e- 2
#> 3 x2 0.0187 0.0214 0.876 3.83e- 1
#> 4 x3 -0.0295 0.0185 -1.60 1.13e- 1
Created on 2023-01-09 with reprex v2.0.2
system
Closed
3
This topic was automatically closed 21 days after the last reply. New replies are no longer allowed.
If you have a query related to it or one of the replies, start a new topic and refer back with a link.