Constructor for a standardising population used for model outputs

Standardised outputs are outputs from models with covariates, that are defined by marginalising (averaging) over covariate values in a given population, rather than being conditional on a given covariate value.

Usage

standardise_to(new_data)

standardize_to(new_data)

Arguments

new_data

Data frame with covariate values to predict for

Details

Standardised outputs are produced from a Monte Carlo sample from the joint distribution of parameters \(\theta\) and covariate values \(X\), \(p(X,\theta) = p(\theta|X)p(X)\), where \(p(X)\) is defined by the empirical distribution of covariates in the standard population. This joint sample is obtained by concatenating samples of covariate-specific outputs.

Hence applying an output function \(g()\) (such as the transition probability) to this sample produces a sample from the posterior of \(\int g(\theta|X) dX\): the average transition probability (say) for a heterogeneous population.

Examples

nd <- data.frame(sex=c("female","male"))

## gender-specific outputs
qdf(infsim_modelc, new_data = nd) 
#> # A tibble: 4 × 4
#>    from    to      value sex   
#>   <int> <int> <rvar[1d]> <chr> 
#> 1     1     2  0.62 ± NA female
#> 2     1     2  0.70 ± NA male  
#> 3     2     1  3.84 ± NA female
#> 4     2     1  3.84 ± NA male  

## averaged over men and women in the same proportions as are in `nd`
## in this case, `nd` has two rows, so we take a 50/50 average
qdf(infsim_modelc, new_data = standardise_to(nd))
#> # A tibble: 2 × 3
#>    from    to         value
#>   <int> <int>    <rvar[1d]>
#> 1     1     2  0.66 ± 0.058
#> 2     2     1  3.84 ± 0.000