Runs an experiment with minimal parameters.
Arguments
- x
A CalmrExperiment or design
data.frame- outputs
A character vector specifying which outputs to parse and aggregate. Defaults to NULL, in which case all model outputs are parsed/aggregated.
- parse
A logical specifying whether the raw results should be parsed. Default = TRUE.
- aggregate
A logical specifying whether the parsed results should be aggregated. Default = TRUE.
- ...
Arguments passed to other functions
Value
A CalmrExperiment with results.
Examples
# Using a data.frame only (throws warning)
df <- get_design("relative_validity")
run_experiment(df, model = "RW1972")
#> Warning: Using default model parameters.
#> -----------------------------
#> CalmrExperiment with model:
#> RW1972
#> -----------------------------
#> Design:
#> Group P1 R1 P2 R2
#> 1 True 10AB(US)/10AC TRUE 1#A TRUE
#> 2 Pseudo 5AB(US)/5AB/5AC(US)/5AC TRUE 1#A TRUE
#> -----------------------------
#> Parameters:
#> $True
#> $True$alphas
#> A B C US
#> 0.4 0.4 0.4 0.4
#>
#> $True$betas_on
#> A B C US
#> 0.4 0.4 0.4 0.4
#>
#> $True$betas_off
#> A B C US
#> 0.4 0.4 0.4 0.4
#>
#> $True$lambdas
#> A B C US
#> 1 1 1 1
#>
#>
#> $Pseudo
#> $Pseudo$alphas
#> A B C US
#> 0.4 0.4 0.4 0.4
#>
#> $Pseudo$betas_on
#> A B C US
#> 0.4 0.4 0.4 0.4
#>
#> $Pseudo$betas_off
#> A B C US
#> 0.4 0.4 0.4 0.4
#>
#> $Pseudo$lambdas
#> A B C US
#> 1 1 1 1
#>
# Using custom parameters
df <- get_design("relative_validity")
pars <- get_parameters(df, model = "HD2022")
pars$alphas["US"] <- 0.6
run_experiment(df, parameters = pars, model = "HD2022")
#> -----------------------------
#> CalmrExperiment with model:
#> HD2022
#> -----------------------------
#> Design:
#> Group P1 R1 P2 R2
#> 1 True 10AB(US)/10AC TRUE 1#A TRUE
#> 2 Pseudo 5AB(US)/5AB/5AC(US)/5AC TRUE 1#A TRUE
#> -----------------------------
#> Parameters:
#> $True
#> $True$alphas
#> A B C US
#> 0.4 0.4 0.4 0.6
#>
#>
#> $Pseudo
#> $Pseudo$alphas
#> A B C US
#> 0.4 0.4 0.4 0.6
#>
# Using make_experiment, for more iterations
df <- get_design("blocking")
pars <- get_parameters(df, model = "SM2007")
exper <- make_experiment(df,
parameters = pars, model = "SM2007",
iterations = 4
)
run_experiment(exper)
#> -----------------------------
#> CalmrExperiment with model:
#> SM2007
#> -----------------------------
#> Design:
#> Group P1 R1 P2 R2
#> 1 Blocking 10N>(US) FALSE 10NL>(US)/10#L FALSE
#> 2 Control FALSE 10NL>(US)/10#L FALSE
#> -----------------------------
#> Parameters:
#> $Blocking
#> $Blocking$alphas
#> L N US
#> 0.4 0.4 0.4
#>
#> $Blocking$lambdas
#> L N US
#> 1 1 1
#>
#> $Blocking$omegas
#> L N US
#> 0.2 0.2 0.2
#>
#> $Blocking$rhos
#> L N US
#> 1 1 1
#>
#> $Blocking$gammas
#> L N US
#> 1 1 1
#>
#> $Blocking$taus
#> L N US
#> 0.2 0.2 0.2
#>
#> $Blocking$order
#> [1] 1
#>
#>
#> $Control
#> $Control$alphas
#> L N US
#> 0.4 0.4 0.4
#>
#> $Control$lambdas
#> L N US
#> 1 1 1
#>
#> $Control$omegas
#> L N US
#> 0.2 0.2 0.2
#>
#> $Control$rhos
#> L N US
#> 1 1 1
#>
#> $Control$gammas
#> L N US
#> 1 1 1
#>
#> $Control$taus
#> L N US
#> 0.2 0.2 0.2
#>
#> $Control$order
#> [1] 1
#>
# Only parsing the associations in the model, without aggregation
run_experiment(exper, outputs = "associations", aggregate = FALSE)
#> -----------------------------
#> CalmrExperiment with model:
#> SM2007
#> -----------------------------
#> Design:
#> Group P1 R1 P2 R2
#> 1 Blocking 10N>(US) FALSE 10NL>(US)/10#L FALSE
#> 2 Control FALSE 10NL>(US)/10#L FALSE
#> -----------------------------
#> Parameters:
#> $Blocking
#> $Blocking$alphas
#> L N US
#> 0.4 0.4 0.4
#>
#> $Blocking$lambdas
#> L N US
#> 1 1 1
#>
#> $Blocking$omegas
#> L N US
#> 0.2 0.2 0.2
#>
#> $Blocking$rhos
#> L N US
#> 1 1 1
#>
#> $Blocking$gammas
#> L N US
#> 1 1 1
#>
#> $Blocking$taus
#> L N US
#> 0.2 0.2 0.2
#>
#> $Blocking$order
#> [1] 1
#>
#>
#> $Control
#> $Control$alphas
#> L N US
#> 0.4 0.4 0.4
#>
#> $Control$lambdas
#> L N US
#> 1 1 1
#>
#> $Control$omegas
#> L N US
#> 0.2 0.2 0.2
#>
#> $Control$rhos
#> L N US
#> 1 1 1
#>
#> $Control$gammas
#> L N US
#> 1 1 1
#>
#> $Control$taus
#> L N US
#> 0.2 0.2 0.2
#>
#> $Control$order
#> [1] 1
#>