Welcome to the dev-version of the
library flipscores on CRAN
Set up
To install this github version type (in R):
#if devtools is not installed yet:
# install.packages("devtools")
library(devtools)
install_github("livioivil/flipscores")
Some examples
library(flipscores)
set.seed(1)
x=(rep(0:1,20))
D=data.frame(y=rbinom(40,1,.25+x*.5),x=x,
z=rnorm(40),id=rep(1:20,each=2))
mod_par=glm(y~x*z,data=D,family = binomial)
summary(mod_par)
#>
#> Call:
#> glm(formula = y ~ x * z, family = binomial, data = D)
#>
#> Deviance Residuals:
#> Min 1Q Median 3Q Max
#> -1.9510 -0.7475 0.5615 0.7202 2.1981
#>
#> Coefficients:
#> Estimate Std. Error z value Pr(>|z|)
#> (Intercept) -0.7001 0.5195 -1.348 0.17778
#> x 2.1557 0.7862 2.742 0.00611 **
#> z -1.1320 0.8147 -1.389 0.16470
#> x:z 1.5070 1.0329 1.459 0.14456
#> ---
#> Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
#>
#> (Dispersion parameter for binomial family taken to be 1)
#>
#> Null deviance: 55.051 on 39 degrees of freedom
#> Residual deviance: 41.830 on 36 degrees of freedom
#> AIC: 49.83
#>
#> Number of Fisher Scoring iterations: 4
mod_par
#>
#> Call: glm(formula = y ~ x * z, family = binomial, data = D)
#>
#> Coefficients:
#> (Intercept) x z x:z
#> -0.7001 2.1557 -1.1320 1.5070
#>
#> Degrees of Freedom: 39 Total (i.e. Null); 36 Residual
#> Null Deviance: 55.05
#> Residual Deviance: 41.83 AIC: 49.83
mod=glm_flipscores(y~x*z,data=D,family = binomial,score_type = "ortho")
summary(mod)
#>
#> Flip Score Test:
#> score_type = orthogonalized
#> n_flips= 1000
#>
#> Call:
#> glm(formula = y ~ x * z, family = binomial, data = D, x = TRUE)
#>
#> Deviance Residuals:
#> Min 1Q Median 3Q Max
#> -1.9510 -0.7475 0.5615 0.7202 2.1981
#>
#> Coefficients:
#> Estimate Score Std. Error z value Pr(>|z|)
#> (Intercept) -0.70008 -0.06760 0.05194 -1.301 0.202
#> x 2.15567 0.10335 0.04041 2.557 0.008 **
#> z -1.13197 -0.05254 0.04660 -1.127 0.302
#> x:z 1.50703 0.04013 0.03397 1.181 0.258
#> ---
#> Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
#>
#> (Dispersion parameter for binomial family taken to be 1)
#>
#> Null deviance: 55.051 on 39 degrees of freedom
#> Residual deviance: 41.830 on 36 degrees of freedom
#> AIC: 49.83
#>
#> Number of Fisher Scoring iterations: 4
print(mod)
#>
#> Flip Score Test:
#> score_type = orthogonalized
#> n_flips= 1000
#> Call: glm(formula = y ~ x * z, family = binomial, data = D, x = TRUE)
#>
#> Coefficients:
#> (Intercept) x z x:z
#> -0.7000829 2.1556675 -1.1319694 1.5070293
mod=glm_flipscores(y~x*z,data=D,family = binomial,score_type = "ortho",
id=D$id)
summary(mod)
#>
#> Flip Score Test:
#> score_type = orthogonalized
#> n_flips= 1000
#>
#> Call:
#> glm(formula = y ~ x * z, family = binomial, data = D, x = TRUE)
#>
#> Deviance Residuals:
#> Min 1Q Median 3Q Max
#> -1.9510 -0.7475 0.5615 0.7202 2.1981
#>
#> Coefficients:
#> Estimate Score Std. Error z value Pr(>|z|)
#> (Intercept) -0.70008 -0.13520 0.10374 -1.303 0.210
#> x 2.15567 0.20670 0.08413 2.457 0.020 *
#> z -1.13197 -0.10507 0.08889 -1.182 0.250
#> x:z 1.50703 0.08026 0.06643 1.208 0.226
#> ---
#> Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
#>
#> (Dispersion parameter for binomial family taken to be 1)
#>
#> Null deviance: 55.051 on 39 degrees of freedom
#> Residual deviance: 41.830 on 36 degrees of freedom
#> AIC: 49.83
#>
#> Number of Fisher Scoring iterations: 4References
Bug reports
If you encounter a bug, please file a reprex (minimal reproducible example) on github.