Logistic Model Predictions
predict_glm.Rdstats::predict doesn't offer confidence intervals. This
function will add the Wald confidence limits on the logit (log-odds) scale or
back-transformed to the probability scale.
Usage
predict_glm(
object,
newdata = NULL,
type = c("link", "response"),
conf.level = 0.95
)Arguments
- object
A glm object with family
binomialand link functionlogit.- newdata
optionally, a data frame in which to look for variables with which to predict. If omitted, the fitted linear predictors are used.
- type
The type of prediction required. The default, "link", is on the scale of the linear predictors; the alternative "response" is on the scale of the response variable. Thus for a default binomial model the default predictions are of log-odds (probabilities on logit scale) and type = "response" gives the predicted probabilities.
- conf.level
The confidence level to use.
See also
stats::predict
Examples
#----------------------------------------------------------------------------
# predict_glm() examples.
#----------------------------------------------------------------------------
library(bkstat)
mod <- glm(
formula = am ~ disp + vs,
data = mtcars,
family = binomial
)
predict_glm(
object = mod,
type = "link"
)
#> estimate lower_ci upper_ci
#> Mazda RX4 4.2484820 -0.18067453 8.677638439
#> Mazda RX4 Wag 4.2484820 -0.18067453 8.677638439
#> Datsun 710 0.6679392 -0.73690578 2.072784254
#> Hornet 4 Drive -5.1480090 -9.67954180 -0.616476198
#> Hornet Sportabout -3.5061157 -6.23169857 -0.780532829
#> Valiant -3.8685004 -7.42825690 -0.308743878
#> Duster 360 -3.5061157 -6.23169857 -0.780532829
#> Merc 240D -0.8325754 -2.42820559 0.763054772
#> Merc 230 -0.6038148 -2.11024449 0.902614938
#> Merc 280 -1.6429309 -3.66147474 0.375613011
#> Merc 280C -1.6429309 -3.66147474 0.375613011
#> Merc 450SE -0.2414301 -1.97802794 1.495167763
#> Merc 450SL -0.2414301 -1.97802794 1.495167763
#> Merc 450SLC -0.2414301 -1.97802794 1.495167763
#> Cadillac Fleetwood -7.8486904 -13.74862138 -1.948759387
#> Lincoln Continental -7.3834145 -12.92425527 -1.842573781
#> Chrysler Imperial -6.6079548 -11.55571020 -1.660199320
#> Fiat 128 1.8039878 -0.06430272 3.672278303
#> Honda Civic 1.9203068 -0.01377675 3.854390267
#> Toyota Corolla 2.0986625 0.05942390 4.137901103
#> Toyota Corona 0.1987861 -1.15676133 1.554333492
#> Dodge Challenger -1.8776502 -3.74648931 -0.008811068
#> AMC Javelin -1.3348284 -3.05177958 0.382122872
#> Camaro Z28 -3.1183858 -5.60401643 -0.632755203
#> Pontiac Firebird -5.0570352 -8.85023091 -1.263839544
#> Fiat X1-9 1.7923559 -0.06948770 3.654199489
#> Porsche 914-2 5.7877696 0.18848499 11.387054179
#> Lotus Europa 1.1681108 -0.39213164 2.728353208
#> Ford Pantera L -3.1571588 -5.66606172 -0.648255885
#> Ferrari Dino 4.8300768 -0.03700294 9.697156491
#> Maserati Bora -1.2185094 -2.91590041 0.478881627
#> Volvo 142E 0.1638904 -1.19214795 1.519928733
predict_glm(
object = mod,
type = "response"
)
#> estimate lower_ci upper_ci
#> Mazda RX4 0.9859153084 4.549538e-01 0.9998297
#> Mazda RX4 Wag 0.9859153084 4.549538e-01 0.9998297
#> Datsun 710 0.6610415664 3.236811e-01 0.8882297
#> Hornet 4 Drive 0.0057773906 6.254625e-05 0.3505833
#> Hornet Sportabout 0.0291387207 1.962251e-03 0.3142051
#> Valiant 0.0204622232 5.938695e-04 0.4234214
#> Duster 360 0.0291387207 1.962251e-03 0.3142051
#> Merc 240D 0.3031007892 8.104701e-02 0.6820166
#> Merc 230 0.3534714185 1.081051e-01 0.7114866
#> Merc 280 0.1620666540 2.505092e-02 0.5928146
#> Merc 280C 0.1620666540 2.505092e-02 0.5928146
#> Merc 450SE 0.4399339584 1.215292e-01 0.8168527
#> Merc 450SL 0.4399339584 1.215292e-01 0.8168527
#> Merc 450SLC 0.4399339584 1.215292e-01 0.8168527
#> Cadillac Fleetwood 0.0003901105 1.069176e-06 0.1246887
#> Lincoln Continental 0.0006210892 2.438182e-06 0.1367472
#> Chrysler Imperial 0.0013477706 9.581084e-06 0.1597352
#> Fiat 128 0.8586336735 4.839299e-01 0.9752116
#> Honda Civic 0.8721726372 4.965559e-01 0.9792530
#> Toyota Corolla 0.8907731132 5.148516e-01 0.9842943
#> Toyota Corona 0.5495335136 2.392563e-01 0.8255387
#> Dodge Challenger 0.1326590106 2.305631e-02 0.4977972
#> AMC Javelin 0.2083618200 4.514071e-02 0.5943850
#> Camaro Z28 0.0423551968 3.669526e-03 0.3468861
#> Pontiac Firebird 0.0063241511 1.433281e-04 0.2203136
#> Fiat X1-9 0.8572158735 4.826351e-01 0.9747708
#> Porsche 914-2 0.9969445540 5.469822e-01 0.9999887
#> Lotus Europa 0.7628033616 4.032043e-01 0.9386791
#> Ford Pantera L 0.0408101262 3.449530e-03 0.3433827
#> Ferrari Dino 0.9920773613 4.907503e-01 0.9999385
#> Maserati Bora 0.2281988764 5.137312e-02 0.6174838
#> Volvo 142E 0.5408811326 2.328750e-01 0.8205280
predict_glm(
object = mod,
newdata = data.frame(disp = c(150, 300), vs = c(0, 0)),
type = "link"
)
#> disp vs estimate lower_ci upper_ci
#> 1 150 0 4.636212 -0.08422335 9.3566470
#> 2 300 0 -1.179736 -2.87171455 0.5122417
predict_glm(
object = mod,
newdata = data.frame(disp = c(150, 300), vs = c(0, 0)),
type = "response"
)
#> disp vs estimate lower_ci upper_ci
#> 1 150 0 0.9903987 0.47895660 0.9999136
#> 2 300 0 0.2350996 0.05356966 0.6253318