Skip to contents

[Stable]

LogisticNormal is the class for the usual logistic regression model with a bivariate normal prior on the intercept and slope.

Usage

LogisticNormal(mean, cov, ref_dose = 1)

.DefaultLogisticNormal()

Arguments

mean

(numeric)
the prior mean vector.

cov

(matrix)
the prior covariance matrix. The precision matrix prec is internally calculated as an inverse of cov.

ref_dose

(number)
the reference dose \(x*\) (strictly positive number).

Details

The covariate is the natural logarithm of the dose \(x\) divided by the reference dose \(x*\), i.e.: $$logit[p(x)] = alpha0 + alpha1 * log(x/x*),$$ where \(p(x)\) is the probability of observing a DLT for a given dose \(x\). The prior $$(alpha0, alpha1) ~ Normal(mean, cov).$$

Note

Typically, end users will not use the .DefaultLogisticNormal() function.

Examples

# Define the dose-grid.
empty_data <- Data(doseGrid = c(1, 3, 5, 10, 15, 20, 25, 40, 50, 80, 100))

my_model <- LogisticNormal(
  mean = c(-0.85, 1),
  cov = matrix(c(1, -0.5, -0.5, 1), nrow = 2)
)

my_options <- McmcOptions(burnin = 10, step = 2, samples = 100)

samples <- mcmc(empty_data, my_model, my_options)
samples
#> An object of class "Samples"
#> Slot "data":
#> $alpha0
#>   [1] -0.50550773 -2.55233435 -1.62405124  0.35780922 -0.71965501  0.42837376
#>   [7] -0.08029425 -0.58437214 -0.88479483 -0.32646873  1.12395254 -0.80328044
#>  [13]  1.06255817 -1.07022490 -0.77342373 -2.52242958 -0.08640770 -0.09028514
#>  [19]  0.27102000 -0.12763081 -1.84685404  0.16925267 -0.22788287 -0.23416974
#>  [25] -0.37145776 -0.90900729 -0.69782063 -2.21460813 -1.77632471 -0.28655472
#>  [31]  0.56853490  0.68305028 -0.90359113 -0.67028281  1.16833165 -1.72144509
#>  [37] -0.24678460 -0.84819388 -0.40834143 -1.05269868 -0.61575612 -0.34734612
#>  [43] -1.39806749 -0.75272103 -0.82192865 -0.70109279  0.16498642 -2.41225549
#>  [49]  1.52832686 -0.51352820 -0.29846433 -3.08949461 -1.34449695 -1.20832138
#>  [55] -1.01918009  0.31581703 -0.93025604 -0.99585170 -0.47280368 -1.76770430
#>  [61]  0.77371410 -2.19797147  0.38874883 -1.56536338  0.15174317  0.03434288
#>  [67] -0.90893331 -0.75234691 -0.76644088 -1.64058341  0.26392064 -1.44312871
#>  [73]  0.63325730 -0.46814029 -0.53934425 -1.75198155 -1.84401974 -1.09249881
#>  [79] -0.80416992 -0.36417688  0.91957942 -0.95050079 -1.48272076 -0.89856651
#>  [85] -1.65551331  0.18251343  0.21480312 -2.20581665 -1.38941597  0.60676792
#>  [91] -1.64479794 -1.22598557 -0.88188276 -0.52800028  0.50303557 -1.27020074
#>  [97] -0.29156016 -3.22252144 -2.21374260 -0.44401091
#> 
#> $alpha1
#>   [1]  1.35392326  3.57007773  2.00166917 -0.27989689  1.46173968 -0.02520215
#>   [7] -0.47914355  2.31133587  0.71806771  0.94885673  0.88481739  0.40003345
#>  [13] -0.13586279  0.95553481  1.92250290  2.06473039 -0.16739264  1.95839353
#>  [19] -0.03466121  1.93838678  2.90910308  0.30451249  0.78789480  1.34938286
#>  [25]  1.84485504  1.97901784  1.69092119  0.82372308  1.97871807  1.23489737
#>  [31]  2.11335459  2.14052299  1.66916985  0.73127349 -0.59483911  1.88080188
#>  [37]  0.63126434  0.20001458  1.46286022  0.00823751  0.65059328  0.76880537
#>  [43]  1.04392131  1.45747260  3.03984767  0.49615867  1.13007542  1.30525039
#>  [49]  0.57222840  1.26025039 -0.08752278  2.32178278  0.65647257  2.08143620
#>  [55]  0.07747305 -1.21434781  1.16135312  1.42945982  1.49168929  1.60108182
#>  [61] -0.75602006  1.43485755  0.54068667  0.55333639  0.96676839  0.66455201
#>  [67]  1.69191646  3.52197760  2.31230423  0.83163427  0.25762222  3.91043577
#>  [73]  1.09832107  1.57423620  0.41283559  1.85250918  1.71835268  1.06294806
#>  [79]  1.29546397  0.54876826  1.42400757  0.35076745  0.34576507  1.20163670
#>  [85]  0.75506254  0.38849803  1.85147485  2.23275006 -0.18297486  0.87974582
#>  [91]  1.66780454  1.91597971  1.16490812 -0.06291770  0.17579833  0.29063930
#>  [97] -0.01257664  0.21809149  1.75031831  2.53735671
#> 
#> 
#> Slot "options":
#> An object of class "McmcOptions"
#> Slot "iterations":
#> [1] 210
#> 
#> Slot "burnin":
#> [1] 10
#> 
#> Slot "step":
#> [1] 2
#> 
#> Slot "rng_kind":
#> [1] NA
#> 
#> Slot "rng_seed":
#> [1] NA
#> 
#>