Default model priors
Usage
epict_priors(individual_variation = c(0, 0.05), include_descriptive = FALSE)Arguments
- individual_variation
A vector of length two specifying the default mean and standard deviation of individual-level variation.
- include_descriptive
Logical, defaults
FALSE. Should an extended list of priors be returned including transformed parameters. This is useful when transforming model output and for user exploration.
See also
Functions used for modelling
epict_convert_to_list(),
epict_inits(),
epict_model()
Examples
epict_priors()
#> variable name
#> 1: t_inf Time between infection and first positive test
#> 2: c_thres Ct value at clearance of infection
#> 3: c_p Ct value at peak
#> 4: t_p Time at peak Ct
#> 5: c_s Ct value at switch
#> 6: t_s Time at switch Ct
#> 7: t_clear Time at clearance of infection
#> 8: inc_mean Incubation period (log) mean
#> 9: inc_sd Incubation period (log) standard deviation
#> 10: sigma Observation standard deviation
#> 11: lkj Lewandowski-Kurowicka-Joe distribution
#> detail
#> 1: Offset by onset if available
#> 2: Offset by user specified limit of Ct detection
#> 3: Logit scale offset by latent limit of detection/Ct at switch
#> 4: Log scale
#> 5: Logit scale offset by latent limit of detection
#> 6: Log scale relative to the time at peak Ct
#> 7: Relative to time at peak and switch C
#> 8: Parameterised as a log-normal distribution
#> 9: Parameterised as a log-normal distribution
#> 10: Assuming a normal error model
#> 11: Prior used for the individual-level correlation matrix. There is \n one parameter which at 1 represents a uniform prior. Smaller values indicate strong correlations, and larger values weaker correlations
#> distribution intercept_mean
#> 1: Normal (truncated by onset or 0) 5.00
#> 2: Normal (truncated by user specified limit of detection) 10.00
#> 3: Normal 0.00
#> 4: Normal 1.61
#> 5: Normal 0.00
#> 6: Normal 1.61
#> 7: Normal 2.30
#> 8: Normal 1.62
#> 9: Zero truncated normal 0.42
#> 10: Zero truncated normal 2.00
#> 11: Lewandowski-Kurowicka-Joe (LKJ) distribution 1.00
#> intercept_sd individual_variation_mean individual_variation_sd
#> 1: 5.00 NA NA
#> 2: 10.00 NA NA
#> 3: 1.00 0 0.05
#> 4: 0.50 0 0.05
#> 5: 1.00 0 0.05
#> 6: 0.50 0 0.05
#> 7: 0.50 0 0.05
#> 8: 0.06 NA NA
#> 9: 0.07 NA NA
#> 10: 2.00 NA NA
#> 11: NA NA NA
# Also include descriptive parameters
epict_priors(include_descriptive = TRUE)
#> variable name
#> 1: t_inf Time between infection and first positive test
#> 2: c_thres Ct value at clearance of infection
#> 3: c_p Ct value at peak
#> 4: t_p Time at peak Ct
#> 5: c_s Ct value at switch
#> 6: t_s Time at switch Ct
#> 7: t_clear Time at clearance of infection
#> 8: inc_mean Incubation period (log) mean
#> 9: inc_sd Incubation period (log) standard deviation
#> 10: sigma Observation standard deviation
#> 11: lkj Lewandowski-Kurowicka-Joe distribution
#> 12: ct_shift Ct intercept adjustment
#> 13: ct_scale Ct multiplicative adjustment
#> 14: nat_inc_mean Incubation period (mean)
#> 15: nat_inc_sd Incubation period (standard deviation)
#> detail
#> 1: Offset by onset if available
#> 2: Offset by user specified limit of Ct detection
#> 3: Logit scale offset by latent limit of detection/Ct at switch
#> 4: Log scale
#> 5: Logit scale offset by latent limit of detection
#> 6: Log scale relative to the time at peak Ct
#> 7: Relative to time at peak and switch C
#> 8: Parameterised as a log-normal distribution
#> 9: Parameterised as a log-normal distribution
#> 10: Assuming a normal error model
#> 11: Prior used for the individual-level correlation matrix. There is \n one parameter which at 1 represents a uniform prior. Smaller values indicate strong correlations, and larger values weaker correlations
#> 12: Cycle threshold intercept adjustment with an intercept set to 0 (i. no \n adjustment. This is adjusted using `adjustment_formula()`
#> 13: Cycle threshold gradient adjustment with an intercept set to 1 (i. no \n adjustment. This is adjusted using `adjustment_formula()`
#> 14: Incubation period mean on the natural scale
#> 15: Incubation period standard deviation on the natural scale
#> distribution intercept_mean
#> 1: Normal (truncated by onset or 0) 5.00
#> 2: Normal (truncated by user specified limit of detection) 10.00
#> 3: Normal 0.00
#> 4: Normal 1.61
#> 5: Normal 0.00
#> 6: Normal 1.61
#> 7: Normal 2.30
#> 8: Normal 1.62
#> 9: Zero truncated normal 0.42
#> 10: Zero truncated normal 2.00
#> 11: Lewandowski-Kurowicka-Joe (LKJ) distribution 1.00
#> 12: NA
#> 13: NA
#> 14: NA
#> 15: NA
#> intercept_sd individual_variation_mean individual_variation_sd
#> 1: 5.00 NA NA
#> 2: 10.00 NA NA
#> 3: 1.00 0 0.05
#> 4: 0.50 0 0.05
#> 5: 1.00 0 0.05
#> 6: 0.50 0 0.05
#> 7: 0.50 0 0.05
#> 8: 0.06 NA NA
#> 9: 0.07 NA NA
#> 10: 2.00 NA NA
#> 11: NA NA NA
#> 12: NA NA NA
#> 13: NA NA NA
#> 14: NA NA NA
#> 15: NA NA NA