The psfmi package contains a function to conduct stability analyses of logistic, Cox and Mixed models. With this function the stability of models selected and the bootstrap inclusion frequency can be evaluated. For logistic and Cox regression models bootstrapping is than used and for Multilevel models cluster bootstrapping.
library(psfmi)
psfmi_res <- psfmi_lr(data=lbpmilr, nimp=5, impvar="Impnr", Outcome="Chronic",
predictors=c("Gender", "Smoking", "JobControl", "Age", "Duration",
"JobDemands", "SocialSupport"), p.crit =0.157, method="MPR",
cat.predictors = "Satisfaction", int.predictors = c("Gender:JobControl"),
direction = "BW")
#> Removed at Step 1 is - Smoking
#> Removed at Step 2 is - Duration
#> Removed at Step 3 is - JobDemands
#> Removed at Step 4 is - factor(Satisfaction)
#> Removed at Step 5 is - Gender*JobControl
#> Removed at Step 6 is - JobControl
#> Removed at Step 7 is - SocialSupport
#> Removed at Step 8 is - Age
#>
#> Selection correctly terminated,
#> No more variables removed from the model
stab_res <- psfmi_stab(psfmi_res, direction="FW", start_model = TRUE,
boot_method = "single", nboot=20, p.crit=0.05)
#>
#> Boot 1
#>
#> Selection correctly terminated,
#> No new variables entered the model
#>
#> Boot 2
#> Entered at Step 1 is - Gender
#>
#> Selection correctly terminated,
#> No new variables entered the model
#>
#> Boot 3
#> Entered at Step 1 is - Gender
#>
#> Selection correctly terminated,
#> No new variables entered the model
#>
#> Boot 4
#>
#> Selection correctly terminated,
#> No new variables entered the model
#>
#> Boot 5
#>
#> Selection correctly terminated,
#> No new variables entered the model
#>
#> Boot 6
#>
#> Selection correctly terminated,
#> No new variables entered the model
#>
#> Boot 7
#>
#> Selection correctly terminated,
#> No new variables entered the model
#>
#> Boot 8
#>
#> Selection correctly terminated,
#> No new variables entered the model
#>
#> Boot 9
#>
#> Selection correctly terminated,
#> No new variables entered the model
#>
#> Boot 10
#> Entered at Step 1 is - Age
#>
#> Selection correctly terminated,
#> No new variables entered the model
#>
#> Boot 11
#> Entered at Step 1 is - Gender
#>
#> Selection correctly terminated,
#> No new variables entered the model
#>
#> Boot 12
#>
#> Selection correctly terminated,
#> No new variables entered the model
#>
#> Boot 13
#>
#> Selection correctly terminated,
#> No new variables entered the model
#>
#> Boot 14
#>
#> Selection correctly terminated,
#> No new variables entered the model
#>
#> Boot 15
#> Entered at Step 1 is - Age
#>
#> Selection correctly terminated,
#> No new variables entered the model
#>
#> Boot 16
#> Entered at Step 1 is - Gender
#>
#> Selection correctly terminated,
#> No new variables entered the model
#>
#> Boot 17
#>
#> Selection correctly terminated,
#> No new variables entered the model
#>
#> Boot 18
#> Entered at Step 1 is - Gender
#>
#> Selection correctly terminated,
#> No new variables entered the model
#>
#> Boot 19
#>
#> Selection correctly terminated,
#> No new variables entered the model
#>
#> Boot 20
#>
#> Selection correctly terminated,
#> No new variables entered the model
stab_res$bif
#> Gender Smoking JobControl Age Duration JobDemands SocialSupport
#> boot 1 0 0 0 0 0 0 0
#> boot 2 1 0 0 0 0 0 0
#> boot 3 1 0 0 0 0 0 0
#> boot 4 0 0 0 0 0 0 0
#> boot 5 0 0 0 0 0 0 0
#> boot 6 0 0 0 0 0 0 0
#> boot 7 0 0 0 0 0 0 0
#> boot 8 0 0 0 0 0 0 0
#> boot 9 0 0 0 0 0 0 0
#> boot 10 0 0 0 1 0 0 0
#> boot 11 1 0 0 0 0 0 0
#> boot 12 0 0 0 0 0 0 0
#> boot 13 0 0 0 0 0 0 0
#> boot 14 0 0 0 0 0 0 0
#> boot 15 0 0 0 1 0 0 0
#> boot 16 1 0 0 0 0 0 0
#> boot 17 0 0 0 0 0 0 0
#> boot 18 1 0 0 0 0 0 0
#> boot 19 0 0 0 0 0 0 0
#> boot 20 0 0 0 0 0 0 0
#> factor(Satisfaction) Gender*JobControl
#> boot 1 0 0
#> boot 2 0 0
#> boot 3 0 0
#> boot 4 0 0
#> boot 5 0 0
#> boot 6 0 0
#> boot 7 0 0
#> boot 8 0 0
#> boot 9 0 0
#> boot 10 0 0
#> boot 11 0 0
#> boot 12 0 0
#> boot 13 0 0
#> boot 14 0 0
#> boot 15 0 0
#> boot 16 0 0
#> boot 17 0 0
#> boot 18 0 0
#> boot 19 0 0
#> boot 20 0 0
stab_res$bif_perc
#> Gender Smoking JobControl
#> 25 0 0
#> Age Duration JobDemands
#> 10 0 0
#> SocialSupport factor(Satisfaction) Gender*JobControl
#> 0 0 0
stab_res$model_stab
#> Gender Smoking JobControl Age Duration JobDemands SocialSupport
#> 1 0 0 0 0 0 0 0
#> 2 1 0 0 0 0 0 0
#> 3 0 0 0 1 0 0 0
#> factor(Satisfaction) Gender*JobControl freq bif_pat_perc
#> 1 0 0 13 65
#> 2 0 0 5 25
#> 3 0 0 2 10Back to Examples
library(psfmi)
pool_lr <- psfmi_coxr(formula = Surv(Time, Status) ~ Pain + factor(Satisfaction) + rcs(Tampascale,3) +
Radiation + Radiation*factor(Satisfaction) + Age + Duration + Previous +
Radiation*rcs(Tampascale, 3), data=lbpmicox, p.crit = 0.157, direction="FW",
nimp=5, impvar="Impnr", keep.predictors = NULL, method="D1")
#> Entered at Step 1 is - Pain
#> Entered at Step 2 is - Duration
#> Entered at Step 3 is - rcs(Tampascale,3)
#>
#> Selection correctly terminated,
#> No new variables entered the model
pool_lr$RR_Model
#> NULL
pool_lr$multiparm
#> $`Step 0 - selected - Pain`
#> p-value D1
#> Pain 0.002972173
#> Radiation 0.181042842
#> Age 0.351345242
#> Duration 0.044028841
#> Previous 0.714681296
#> factor(Satisfaction) 0.708284253
#> rcs(Tampascale,3) 0.032223680
#> factor(Satisfaction)*Radiation 0.712860014
#> rcs(Tampascale,3)*Radiation 0.452891382
#>
#> $`Step 1 - selected - Duration`
#> p-value D1
#> Radiation 0.31938607
#> Age 0.22333023
#> Duration 0.04853205
#> Previous 0.54425893
#> factor(Satisfaction) 0.76439993
#> rcs(Tampascale,3) 0.07594972
#> factor(Satisfaction)*Radiation 0.66328781
#> rcs(Tampascale,3)*Radiation 0.44642240
#>
#> $`Step 2 - selected - rcs(Tampascale,3)`
#> p-value D1
#> Radiation 0.32679294
#> Age 0.40528789
#> Previous 0.44216497
#> factor(Satisfaction) 0.81050733
#> rcs(Tampascale,3) 0.06066281
#> factor(Satisfaction)*Radiation 0.70440131
#> rcs(Tampascale,3)*Radiation 0.35955920
stab_res <- psfmi_stab(pool_lr, direction="FW", start_model = TRUE,
boot_method = "single", nboot=20, p.crit=0.05)
#>
#> Boot 1
#> Entered at Step 1 is - Pain
#>
#> Selection correctly terminated,
#> No new variables entered the model
#>
#> Boot 2
#> Entered at Step 1 is - Pain
#>
#> Selection correctly terminated,
#> No new variables entered the model
#>
#> Boot 3
#>
#> Selection correctly terminated,
#> No new variables entered the model
#>
#> Boot 4
#>
#> Selection correctly terminated,
#> No new variables entered the model
#>
#> Boot 5
#> Entered at Step 1 is - Pain
#>
#> Selection correctly terminated,
#> No new variables entered the model
#>
#> Boot 6
#> Entered at Step 1 is - Pain
#>
#> Selection correctly terminated,
#> No new variables entered the model
#>
#> Boot 7
#>
#> Selection correctly terminated,
#> No new variables entered the model
#>
#> Boot 8
#> Entered at Step 1 is - Pain
#>
#> Selection correctly terminated,
#> No new variables entered the model
#>
#> Boot 9
#>
#> Selection correctly terminated,
#> No new variables entered the model
#>
#> Boot 10
#>
#> Selection correctly terminated,
#> No new variables entered the model
#>
#> Boot 11
#> Entered at Step 1 is - Pain
#>
#> Selection correctly terminated,
#> No new variables entered the model
#>
#> Boot 12
#> Entered at Step 1 is - Pain
#>
#> Selection correctly terminated,
#> No new variables entered the model
#>
#> Boot 13
#>
#> Selection correctly terminated,
#> No new variables entered the model
#>
#> Boot 14
#>
#> Selection correctly terminated,
#> No new variables entered the model
#>
#> Boot 15
#> Entered at Step 1 is - rcs(Tampascale,3)
#>
#> Selection correctly terminated,
#> No new variables entered the model
#>
#> Boot 16
#>
#> Selection correctly terminated,
#> No new variables entered the model
#>
#> Boot 17
#> Entered at Step 1 is - Pain
#>
#> Selection correctly terminated,
#> No new variables entered the model
#>
#> Boot 18
#>
#> Selection correctly terminated,
#> No new variables entered the model
#>
#> Boot 19
#> Entered at Step 1 is - Pain
#>
#> Selection correctly terminated,
#> No new variables entered the model
#>
#> Boot 20
#> Entered at Step 1 is - Pain
#>
#> Selection correctly terminated,
#> No new variables entered the model
stab_res$bif
#> Pain Radiation Age Duration Previous factor(Satisfaction)
#> boot 1 1 0 0 0 0 0
#> boot 2 1 0 0 0 0 0
#> boot 3 0 0 0 0 0 0
#> boot 4 0 0 0 0 0 0
#> boot 5 1 0 0 0 0 0
#> boot 6 1 0 0 0 0 0
#> boot 7 0 0 0 0 0 0
#> boot 8 1 0 0 0 0 0
#> boot 9 0 0 0 0 0 0
#> boot 10 0 0 0 0 0 0
#> boot 11 1 0 0 0 0 0
#> boot 12 1 0 0 0 0 0
#> boot 13 0 0 0 0 0 0
#> boot 14 0 0 0 0 0 0
#> boot 15 0 0 0 0 0 0
#> boot 16 0 0 0 0 0 0
#> boot 17 1 0 0 0 0 0
#> boot 18 0 0 0 0 0 0
#> boot 19 1 0 0 0 0 0
#> boot 20 1 0 0 0 0 0
#> rcs(Tampascale,3) factor(Satisfaction)*Radiation
#> boot 1 0 0
#> boot 2 0 0
#> boot 3 0 0
#> boot 4 0 0
#> boot 5 0 0
#> boot 6 0 0
#> boot 7 0 0
#> boot 8 0 0
#> boot 9 0 0
#> boot 10 0 0
#> boot 11 0 0
#> boot 12 0 0
#> boot 13 0 0
#> boot 14 0 0
#> boot 15 1 0
#> boot 16 0 0
#> boot 17 0 0
#> boot 18 0 0
#> boot 19 0 0
#> boot 20 0 0
#> rcs(Tampascale,3)*Radiation
#> boot 1 0
#> boot 2 0
#> boot 3 0
#> boot 4 0
#> boot 5 0
#> boot 6 0
#> boot 7 0
#> boot 8 0
#> boot 9 0
#> boot 10 0
#> boot 11 0
#> boot 12 0
#> boot 13 0
#> boot 14 0
#> boot 15 0
#> boot 16 0
#> boot 17 0
#> boot 18 0
#> boot 19 0
#> boot 20 0
stab_res$bif_perc
#> Pain Radiation
#> 50 0
#> Age Duration
#> 0 0
#> Previous factor(Satisfaction)
#> 0 0
#> rcs(Tampascale,3) factor(Satisfaction)*Radiation
#> 5 0
#> rcs(Tampascale,3)*Radiation
#> 0
stab_res$model_stab
#> Pain Radiation Age Duration Previous factor(Satisfaction) rcs(Tampascale,3)
#> 1 1 0 0 0 0 0 0
#> 2 0 0 0 0 0 0 0
#> 3 0 0 0 0 0 0 1
#> factor(Satisfaction)*Radiation rcs(Tampascale,3)*Radiation freq bif_pat_perc
#> 1 0 0 10 50
#> 2 0 0 9 45
#> 3 0 0 1 5Back to Examples