This function use bootstrapping to conduct hypothesis testing for the partial association coefficients. It directly applies onto the "PAsso" class of object generated by "PAsso".
Arguments
- object
An object of "PAsso" class, which is generated by "PAsso" function.
- bootstrap_rep
The number of bootstrap replications. It may be slow.
- H0
null hypothesis of partial correlation coefficient.
- parallel
logical argument whether conduct parallel for bootstrapping partial association.
Examples
# Import ANES2016 data in "PAsso"
data(ANES2016)
# Parial association:
PAsso_2v <- PAsso(responses = c("PreVote.num", "PID"),
adjustments = c("income.num", "age", "edu.year"),
data = ANES2016)
summary(PAsso_2v, digits=4)
#> --------------------------------------------
#> The partial correlation coefficient matrix:
#>
#> PreVote.num PID
#> PreVote.num 1.0000 0.4492
#> PID 1.0000
#> --------------------------------------------
#> The marginal correlation coefficient matrix:
#>
#> PreVote.num PID
#> PreVote.num 1.0000 0.7059
#> PID 1.0000
#>
#> --------------------------------------------
#> --------------------------------------------
#>
#> The coefficients of fitted models are:
#>
#> PreVote.num PID
#> income.num 0.0005 0.0009*
#> Std. Error 0.0005 0.0004
#> ---
#> age 0.0092*** 0.0048***
#> Std. Error 0.0016 0.0013
#> ---
#> edu.year -0.0798*** -0.0459***
#> Std. Error 0.0117 0.0098
#> ---
#> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
PAsso_2v_test <- test(object = PAsso_2v, bootstrap_rep=20, H0=0, parallel=FALSE)
#> The bootstrapping procedure may be slow when bootstrap_rep is large!
PAsso_2v_test
#> --------------------------------------------
#> The partial association analysis:
#>
#> PreVote.num PID
#> PreVote.num 1.0000 0.4492
#> S.E. 0.006
#> Pr 0.05*
#> ---
#> PID 1.0000
#> S.E.
#> Pr
#> ---
#> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1