Movatterモバイル変換

Generating function for Bernoulli Data.

Description

genlogisticdata() function used mainly internally bylogistictrialsimulator() function to generate data for a two-armclinical trial, experimental and control groups. Can be used to generaterandom trial data.

Usage

genbernoullidata(sample_size, prob1, odds_ratio)

Arguments

Value

genlogisticdata() returns a data frame with columns: 'id', 'treatment',and 'y'.

Examples

samplehistdata <- genbernoullidata(sample_size=60, prob1=0.6, odds_ratio=0.6)samplehistdata

Generating function for Gaussian Data.

Description

gengaussiandata() function used mainly internally bygaussiantrialsimulator() function to generate data for a two-armclinical trial, experimental and control groups. Can be used to generaterandom trial data.

Usage

gengaussiandata(sample_size, mu1, mean_diff, common_sd)

Arguments

Value

gengaussiandata() returns a data frame with columns: 'id', 'treatment',and 'y'.

Examples

samplehistdata <- gengaussiandata(sample_size=60, mu1=25, mean_diff=0, common_sd=3)samplehistdata

Generating function for Lognormal Data.

Description

genlognormaldata() function used mainly internally bylognormaltrialsimulator() andlognormaltrialsimulatornohist() functionsto generate data for a two-arm clinical trial, experimental and control groups.Can be used to generate random trial data.

Usage

genlognormaldata(sample_size, mu1, mean_ratio, common_sd, censor_value)

Arguments

Value

genlognormaldata() returns a data frame with columns: 'id', 'treatment','event_time', and 'status'.

Examples

samplehistdata <- genlognormaldata(sample_size=60, mu1=1.06, mean_ratio=0.6,                                   common_sd=1.25, censor_value=3)samplehistdata

Generating function for Poisson Data.

Description

genpoissondata() function mainly used internally bypoissontrialsimulator() function to generate data for a two-armclinical trial, experimental and control groups. Can be used to generaterandom trial data.

Usage

genpoissondata(sample_size, mu1, mean_ratio)

Arguments

Value

genpoissondata() returns a data frame with columns: 'id', 'treatment',and 'y'.

Examples

samplehistdata <- genpoissondata(sample_size=60, mu1=1, mean_ratio=1.0)samplehistdata

Generating function for Piece-wise Exponential Data.

Description

genpwedata() function used mainly internally bypwetrialsimulator() function to generate data for a two-armclinical trial, experimental and control groups. Can be used to generaterandom trial data.

Usage

genpwedata(sample_size, lambda_vec, hazard_ratio, time_vec, censor_value)

Arguments

Value

genpwedata() returns a data frame with columns: 'id', 'treatment','event_time', 'status', and 'indicator'.

Examples

nvalHC <- 60time.vec <- c(0.3,0.9,1.5,2.1,2.4)lambdaHC.vec <- c(0.19,0.35,0.56,0.47,0.38,0.34)censor.value <- 3SampleHistData <- genpwedata(nvalHC, lambdaHC.vec, 1.0, time.vec, censor.value)SampleHistData

Generating function for Weibull Data.

Description

genweibulldata() function used mainly internally byweibulltrialsimulator() andweibulltrialsimulatornohist() functionsto generate data for a two-arm clinical trial, experimental and control groups.Can be used to generate random trial data.

Usage

genweibulldata(sample_size, scale1, hazard_ratio, common_shape, censor_value)

Arguments

Value

genweibulldata() returns a data frame with columns: 'id', 'treatment','event_time', and 'status'.

Examples

SampleHistData <- genweibulldata(sample_size=60, scale1=2.82487,                                 hazard_ratio=0.6, common_shape=3,                                 censor_value=3)SampleHistData

Two Arm Bayesian Clinical Trial Simulation with Historical Data

Description

historic_sim() returns an S3 object of classbayes_ctd_array, whichwill contain simulation results for power, statistic estimation, bias,variance, and mse as requested by user.

Usage

historic_sim(  trial_reps = 100,  outcome_type = "weibull",  subj_per_arm = c(50, 100, 150, 200, 250),  a0_vals = c(0, 0.33, 0.67, 1),  effect_vals = c(0.6, 1, 1.4),  rand_control_diff = c(0.8, 1, 1.2),  hist_control_data = NULL,  time_vec = NULL,  censor_value = NULL,  alpha = 0.05,  get_var = FALSE,  get_bias = FALSE,  get_mse = FALSE,  seedval = NULL,  quietly = TRUE)

Arguments

Details

The objectbayes_ctd_array has 6 elements: a list containing simulationresults (data), copies of the 4 function argumentssubj_per_arm,a0_vals,effect_vals, andrand_control_diff, and finallyaobjtype value indicating thathistoric_sim() was used. Each element ofdata is a four-dimensional array, where each dimension is determined by thelength of parameterssubj_per_arm,a0_vals,effect_vals, andrand_control_diff. The size ofdata depends on which results arerequested by the user. At a minimum, at least one ofsubj_per_arm,a0_vals,effect_vals, orrand_control_diff must contain atleast 2 values, while the other three must contain at least 1 value. Thedatalist will always contain two elements: an array of power results (power) andan array of estimation results (est). In addition topower andest, data may also contain elementsvar,bias, ormse,depending on the values ofget_var,get_bias, andget_mse. Thevalues returned inest are in the form of hazard ratios, mean ratios, oddsratios, or mean differences depending on the value ofoutcome_type. For aGaussian outcome, the estimation results are differences in group means (experimentalgroup minus control group). For a logistic outcome, the estimation results are oddsratios (experimental group over control group). For lognormal and Poisson outcomes,the estimation results are mean ratios (experimental group over control group). For apiecewise exponential or a Weibull outcome, the estimation results are hazardratios (experimental group over control group). The values returned inbias,var, andmse are on the scale of the values returned inest.

The objectbayes_ctd_array has two primary methods,print() andplot(), for printing and plotting slices of the arrays contained inbayes_ctd_array$data.

As dimensions of the four dimensional array increases, the time required to completethe simulation will increase; however, it will be faster than a similar simulationbased on repeated calls to MCMC routines to analyze each simulated trial.

The meaning of the estimation results, and the test used to generate power results,depends on the outcome used. In all cases, power is based on a two-sided testinvolving a (1-alpha)100% credible interval, where the interval is used to determineif the null hypothesis should be rejected (null value outside of the interval) ornot rejected (null value inside the interval). For a Gaussian outcome, the 95%credible interval is an interval for the difference in group means(experimental group minus control group), and the test determines if 0 is in oroutside of the interval. For a Bernoulli outcome, the 95% credible intervalis an interval for the odds ratio (experimental group over control group),and the test determines if 1 is in or outside of the interval. For a lognormal ora Poisson outcome, the 95% credible interval is an interval for the mean ratio(experimental group over control group), and the test determines if 1 is in oroutside of the interval. Finally, for a piecewise exponential or a Weibull outcome,the 95% credible interval is an interval for the hazard ratio (experimental groupover control group), and the test determines if 1 is in or outside of the interval.

Please refer to the examples for illustration of package use.

Value

historic_sim() returns an S3 object of classbayes_ctd_array.As noted in details, an object of classbayes_ctd_arrayhas 6 elements: alist of simulation results (data), copies of the 4 function argumentssubj_per_arm,a0_vals,effect_vals, andrand_control_diff, and finallyobjtype indicating thathistoric_sim()was used. See details for a discussion about the contents ofdata. Results from the simulation contained in thebayes_ctd_arrayobject can be printed or plotted using theprint() andplot() methods. The results can also be accessed using basic listelement identification and array slicing. For example, to get the 4-dimensionalarray of power results from a simulation, one could use the codebayes_ctd_array$data$power, wherebayes_ctd_array is replacedwith the name of the variable containing thebayes_ctd_array object. Ifone wanted a table of power for sample size by a0, while holding effect equal tothe first considered value and control differences equal to the second consideredvalue, then the code isbayes_ctd_array$data$power[,,1,2], wherebayes_ctd_array is replaced with the name of the variable containing thebayes_ctd_array object.

Examples

#Generate a sample of historical data for use in example.set.seed(2250)SampleHistData <- genweibulldata(sample_size=60, scale1=2.82487,                                 hazard_ratio=0.6, common_shape=3,                                 censor_value=3)histdata <- subset(SampleHistData, subset=(treatment==0))histdata$id <- histdata$id+10000#Run a Weibull simulation, using historic_sim().#For meaningful results, trial_reps needs to be much larger than 2.weibull_test <- historic_sim(trial_reps = 2, outcome_type = "weibull",                             subj_per_arm = c(50, 100, 150),                             a0_vals = c(0, 0.50, 1),                             effect_vals = c(0.6, 1),                             rand_control_diff = c(0.8, 1),                             hist_control_data = histdata, time_vec = NULL,                             censor_value = 3, alpha = 0.05, get_var = TRUE,                             get_bias = TRUE, get_mse = TRUE, seedval=123,                             quietly=TRUE)#Tabulate the simulation results for power.test_table <- print(x=weibull_test, measure="power",                    tab_type="WX|YZ", effect_val=0.6,                    rand_control_diff_val=1.0)print(test_table)#Create a plot of the power simulation results.plot(x=weibull_test, measure="power", tab_type="WX|YZ",     smooth=FALSE, plot_out=TRUE, effect_val=0.6,     rand_control_diff_val=1.0)#Create a plot of the estimated hazard ratio simulation results.plot(x=weibull_test, measure="est", tab_type="WX|YZ",     smooth=FALSE, plot_out=TRUE, effect_val=0.6,     rand_control_diff_val=1.0)#Create a plot of the hazard ratio variance simulation results.plot(x=weibull_test, measure="var", tab_type="WX|YZ",     smooth=FALSE, plot_out=TRUE, effect_val=0.6,     rand_control_diff_val=1.0)#Create a plot of the hazard ratio bias simulation results.plot(x=weibull_test, measure="bias", tab_type="WX|YZ",     smooth=FALSE, plot_out=TRUE, effect_val=0.6,     rand_control_diff_val=1.0)#Create a plot of the hazard ratio mse simulation results.plot(x=weibull_test, measure="mse", tab_type="WX|YZ",     smooth=FALSE, plot_out=TRUE, effect_val=0.6,     rand_control_diff_val=1.0)#Create other power plots using different values for tab_typeplot(x=weibull_test, measure="power", tab_type="XY|WZ",     smooth=FALSE, plot_out=TRUE, subj_per_arm_val=150,     rand_control_diff_val=1.0)plot(x=weibull_test, measure="power", tab_type="XZ|WY",     smooth=FALSE, plot_out=TRUE, subj_per_arm_val=150, effect_val=0.6)plot(x=weibull_test, measure="power", tab_type="YZ|WX",     smooth=FALSE, plot_out=TRUE, subj_per_arm_val=150, a0_val=0.5)plot(x=weibull_test, measure="power", tab_type="WY|XZ",     smooth=FALSE, plot_out=TRUE, rand_control_diff_val=1, a0_val=0.5)plot(x=weibull_test, measure="power", tab_type="WZ|XY",     smooth=FALSE, plot_out=TRUE, effect_val=0.6, a0_val=0.5)#Run Poisson simulation, using historic_sim(), but set two design characteristic# parameters to only 1 value.#Note: historic_sim() can take a while to run.#Generate a sample of historical poisson data for use in example.set.seed(2250)samplehistdata <- genpoissondata(sample_size=60, mu1=1, mean_ratio=1.0)histdata <- subset(samplehistdata, subset=(treatment==0))histdata$id <- histdata$id+10000#For meaningful results, trial_reps needs to be larger than 100.poisson_test <- historic_sim(trial_reps = 100, outcome_type = "poisson",                              subj_per_arm = c(50, 75, 100, 125, 150, 175, 200, 225, 250),                              a0_vals = c(1),                              effect_vals = c(0.6),                              rand_control_diff = c(0.6, 1, 1.6),                              hist_control_data = histdata, time_vec = NULL,                              censor_value = 3, alpha = 0.05, get_var = TRUE,                              get_bias = TRUE, get_mse = TRUE, seedval=123,                              quietly=TRUE)#Tabulate the simulation results for power.test_table <- print(x=poisson_test, measure="power",                    tab_type=NULL)print(test_table)#Create a plot of the power simulation results.plot(x=poisson_test, measure="power", tab_type=NULL,     smooth=FALSE, plot_out=TRUE)#At least one of subj_per_arm, a0_vals, effect_vals, or rand_control_diff#must contain at least 2 values.#Generate a sample of historical lognormal data for use in example.set.seed(2250)samplehistdata <- genlognormaldata(sample_size=60, mu1=1.06, mean_ratio=0.6, common_sd=1.25,                                   censor_value=3)histdata <- subset(samplehistdata, subset=(treatment==0))histdata$id <- histdata$id+10000#Run a Lognormal simulation, using historic_sim().#For meaningful results, trial_reps needs to be larger than 100.lognormal_test <- historic_sim(trial_reps = 100, outcome_type = "lognormal",                               subj_per_arm = c(25,50,75,100,125,150,175,200,225,250),                               a0_vals = c(1.0),                               effect_vals = c(0.6),                               rand_control_diff = c(1.8),                               hist_control_data = histdata, time_vec = NULL,                               censor_value = 3, alpha = 0.05, get_var = TRUE,                               get_bias = TRUE, get_mse = TRUE, seedval=123,                               quietly=TRUE)test_table <- print(x=lognormal_test, measure="power",                    tab_type=NULL)print(test_table)#Create a plot of the power simulation results.plot(x=lognormal_test, measure="power", tab_type=NULL,     smooth=TRUE, plot_out=TRUE)

Plot Data from Two Arm Bayesian Clinical Trial Simulation.

Description

plot.bayes_ctd_array() takes an S3 object of classbayes_ctd_array, andcreates a line plot from a one or two dimensional slice of the data generated by aclinical trial simulation usinghistoric_sim() orsimple_sim(). Theplotted results can be smoothed or unsmoothed.

Usage

## S3 method for class 'bayes_ctd_array'plot(  x = NULL,  measure = "power",  tab_type = "WX|YZ",  smooth = FALSE,  plot_out = TRUE,  subj_per_arm_val = NULL,  a0_val = NULL,  effect_val = NULL,  rand_control_diff_val = NULL,  span = 0.75,  degree = 2,  family = "gaussian",  title = NULL,  ylim = NULL,  ...)

Arguments

Details

If the object of classbayes_ctd_array is created byhistoric_sim(),the functionplot() allows the user to create line plots of user-specified1- or 2- dimensional slices of the simulation results based on slicing codedescribed below. If the object of classbayes_ctd_array is created bysimple_sim(), a basic plot of characteristic by sample size and effect is created.

If the object of classbayes_ctd_array is created bysimple_sim(), thenall four trial characteristics (subj_per_arm_val,a0_vals,effect_val, andrand_control_diff_val) can be ignored as can theparameter defining what type of plot to create through the parametertab_type.A call toplot() will require the user to specify a measure (power, est,var, bias, or mse).

If the object of classbayes_ctd_array is created byhistoric_sim(),when callingplot() the user must specify a measure to plot(power, est, var, bias, or mse) and may be required to specify a plot type throughthetab_type parameter. A plot type,tab_type, will be required if3 of the 4 trial characteristics are equal to a vector of 2 or more values. Thisplot type specification uses the letters W, X, Y, and Z. The letter W representsthe subject per arm dimension. The letter X represents the a0 dimension. Theletter Y represents the effect dimension. The letter Z represents the controldifference dimension. To plot a slice of the 4-dimensional array, these lettersare put into an AB|CD pattern just like inprint(). The two lettersto the right of the vertical bar define which variables are held constant. The twoletters to the left of the vertical bar define which variables are going to show upin the plot. The first letter defines the x-axis variable and the second letterdefines the stratification variable. The result is a plot of power, estimate,variance, bias, or mse by the trial characteristic represented by the first letter.On this plot, one line will be created for each value of the trial characteristicrepresented by the second letter. For example if tab_type equalsWX|YZ,then effect and control differences will be held constant, while sample size will berepresented along the horizontal axis and a0 values will be represented by separatelines. The actual values that are plotted on the y-axis depend on what measure isrequested in the parametermeasure.

• tab_type='WX|YZ', Sample Size by a0

• tab_type='WY|XZ', Sample Size by Effect

• tab_type='WZ|XY', Sample Size by Control Differences

• tab_type='XY|WZ', a0 by Effect

• tab_type='XZ|WY', a0 by Control Differences

• tab_type='YZ|WX', Effect by Control Differences

• tab_type='ZX|WY', Control Differences by a0

• tab_type='XW|YZ', a0 by Sample Size

• tab_type='YW|XZ', Effect by Sample Size

• tab_type='YX|WZ', Effect by a0

• tab_type='ZW|XY', Control Differences by Sample Size

• tab_type='ZY|WX', Control Differences by Effect

It is very important to populate the values ofsubj_per_arm_val,a0_val,effect_val, andrand_control_diff_val correctly giventhe value of tab_type, when the object of classbayes_ctd_array is created byhistoric_sim() and at least 3 of the four parameters have more than onevalue. On, the other hand, if 2 or more of the four parameters have only one value,thensubj_per_arm_val,a0_vals,effect_val,rand_control_diff_val, as well astab_type can be ignored. If the lasttwo letters areYZ, theneffect_val andrand_control_diff_valmust be populated. If the last two letters areXZ, thena0_val andrand_control_diff_val must be populated. If the last two letters areXY,thena0_val andeffect_val must be populated. If the last two lettersareWZ, thensample_val andrand_control_diff_val must bepopulated. If the last two letters areWY, thensample_size_val andeffect_val must be populated. If the last two letters areWX, thensample_size_val anda0_val must be populated.

If the object of classbayes_ctd_array is created bysimple_sim(), theparameterstab_type,subj_per_arm_val,a0_val,effect_val,andrand_control_diff_val are ignored.

Value

plot() returns a plot for a two dimensional array of simulationresults. Ifsmooth isTRUE, then the plot is based on a smoothedversion of the simulation results. Ifsmooth isFALSE, then the plotis based on the raw data from the simulation results. What actually is plotteddepends on the value ofmeasure. Ifplot_out isFALSE, theplot is not created. This option is useful when the user wants a table of smoothedsimulation results but does not want the plot. Smoothing of simulation resultsrequires the length ofsubj_per_arm_val ora0_val oreffect_valorrand_control_diff_val, whichever populates the x-axis on the graph tocontain enough elements to justify the smoothing. No checking occurs todetermine if enough elements are present to justify smoothing.

Examples

#Run a Weibull simulation, using simple_sim().#For meaningful results, trial_reps needs to be much larger than 2.weibull_test <- simple_sim(trial_reps = 2, outcome_type = "weibull",                           subj_per_arm = c(50, 100, 150, 200),                           effect_vals = c(0.6, 1),                           control_parms = c(2.82487,3), time_vec = NULL,                           censor_value = NULL, alpha = 0.05,                           get_var = TRUE, get_bias = TRUE, get_mse = TRUE,                           seedval=123, quietly=TRUE)#Create a plot of the power simulation results.plot(x=weibull_test, measure="power", tab_type=NULL,     smooth=FALSE, plot_out=TRUE, subj_per_arm_val=NULL, a0_val=NULL,     effect_val=NULL, rand_control_diff_val=NULL)#Create a plot of the hazard ratio simulation results.plot(x=weibull_test, measure="est", tab_type=NULL,     smooth=FALSE, plot_out=TRUE, subj_per_arm_val=NULL, a0_val=NULL,     effect_val=NULL, rand_control_diff_val=NULL)#Create a plot of the hazard ratio variance simulation results.plot(x=weibull_test, measure="var", tab_type=NULL,     smooth=FALSE, plot_out=TRUE, subj_per_arm_val=NULL, a0_val=NULL,     effect_val=NULL, rand_control_diff_val=NULL)#Create a plot of the hazard ratio bias simulation results.plot(x=weibull_test, measure="bias", tab_type=NULL,     smooth=FALSE, plot_out=TRUE, subj_per_arm_val=NULL, a0_val=NULL,     effect_val=NULL, rand_control_diff_val=NULL)#Create a plot of the hazard ratio mse simulation results.plot(x=weibull_test, measure="mse", tab_type=NULL,     smooth=FALSE, plot_out=TRUE, subj_per_arm_val=NULL, a0_val=NULL,     effect_val=NULL, rand_control_diff_val=NULL)#Run a second Weibull simulation, using simple_sim() and smooth the plot.#For meaningful results, trial_reps needs to be larger than 100.weibull_test2 <- simple_sim(trial_reps = 100, outcome_type = "weibull",                            subj_per_arm = c(50, 75, 100, 125, 150, 175, 200, 225, 250),                            effect_vals = c(0.6, 1, 1.4),                            control_parms = c(2.82487,3), time_vec = NULL,                            censor_value = NULL, alpha = 0.05, get_var = TRUE,                            get_bias = TRUE, get_mse = TRUE, seedval=123,                            quietly=TRUE)#Create a plot of the power simulation results.plot(x=weibull_test2, measure="power", tab_type=NULL,     smooth=TRUE, plot_out=TRUE, subj_per_arm_val=NULL, a0_val=NULL,     effect_val=NULL, rand_control_diff_val=NULL, span=c(1,1,1))#Run a third weibull simulation, using historic_sim().#Note: historic_sim() can take a while to run.#Generate a sample of historical data for use in example.set.seed(2250)SampleHistData <- genweibulldata(sample_size=60, scale1=2.82487,                                 hazard_ratio=0.6, common_shape=3,                                 censor_value=3)histdata <- subset(SampleHistData, subset=(treatment==0))histdata$id <- histdata$id+10000#For meaningful results, trial_reps needs to be larger than 100.weibull_test3 <- historic_sim(trial_reps = 100, outcome_type = "weibull",                              subj_per_arm = c(50, 100, 150, 200, 250),                              a0_vals = c(0, 0.33, 0.67, 1),                              effect_vals = c(0.6, 1, 1.4),                              rand_control_diff = c(0.8, 1, 1.2),                              hist_control_data = histdata, time_vec = NULL,                              censor_value = 3, alpha = 0.05, get_var = TRUE,                              get_bias = TRUE, get_mse = TRUE, seedval=123,                              quietly=TRUE)#Create a plot of the power simulation results.plot(x=weibull_test3, measure="power", tab_type="WX|YZ",     smooth=FALSE, plot_out=TRUE, effect_val=0.6,     rand_control_diff_val=1.0)#Run a Gaussian simulation, using historic_sim()#Generate a sample of historical Gaussian data for use in example.set.seed(2250)samplehistdata <- gengaussiandata(sample_size=60, mu1=25, mean_diff=0, common_sd=3)histdata <- subset(samplehistdata, subset=(treatment==0))histdata$id <- histdata$id+10000#For meaningful results, trial_reps needs to be larger than 100.gaussian_test <- historic_sim(trial_reps = 100, outcome_type = "gaussian",                             subj_per_arm = c(150),                             a0_vals = c(1.0),                             effect_vals = c(0.15),                             rand_control_diff = c(-4.0,-3.5,-3.0,-2.5,-2.0,                                                   -1.5,-1.0,-0.5,0,0.5,1.0),                             hist_control_data = histdata, time_vec = NULL,                             censor_value = 3, alpha = 0.05, get_var = TRUE,                             get_bias = TRUE, get_mse = TRUE, seedval=123,                             quietly=TRUE)test_table <- print(x=gaussian_test, measure="power",                         tab_type=NULL, effect_val=NULL,                         subj_per_arm_val=NULL)print(test_table)#Create a plot of the power simulation results.plot(x=gaussian_test, measure="power", tab_type=NULL,     smooth=TRUE, plot_out=TRUE, effect_val=NULL,     rand_control_diff_val=NULL)#Generate a sample of historical pwe data for use in example.set.seed(2250)nvalHC <- 60time.vec <- c(0.3,0.9,1.5,2.1,2.4)lambdaHC.vec <- c(0.19,0.35,0.56,0.47,0.38,0.34)censor.value <- 3SampleHistData <- genpwedata(nvalHC, lambdaHC.vec, 1.0, time.vec, censor.value)histdata <- subset(SampleHistData, subset=(treatment==0))histdata$indicator <- 2 #If set to 2, then historical controls will be collapsed with#randomized controls, when time_vec is re-considered and#potentially restructured.  If set to 1, then historical#controls will be treated as a separated cohort when#time_vec is being assessed for restructuring.histdata$id <- histdata$id+10000#Run a pwe simulation, using historic_sim().#For meaningful results, trial_reps needs to be larger than 100.pwe_test <- historic_sim(trial_reps = 100, outcome_type = "pwe",                        subj_per_arm = c(25,50,75,100,125,150,175,200,225,250),                        a0_vals = c(1.0),                        effect_vals = c(0.6),                        rand_control_diff = c(1.8),                        hist_control_data = histdata, time_vec = time.vec,                        censor_value = 3, alpha = 0.05, get_var = TRUE,                        get_bias = TRUE, get_mse = TRUE, seedval=123,                        quietly=TRUE)#Create a plot of the power simulation results.plot(x=pwe_test, measure="power", tab_type=NULL,     smooth=TRUE, plot_out=TRUE, effect_val=NULL,     rand_control_diff_val=NULL)

Print Data from Two Arm Bayesian Clinical Trial Simulation.

Description

print.bayes_ctd_array() takes an S3 object of classbayes_ctd_array, andprints a two dimensional slice from the data generated by a clinical trial simulationusinghistoric_sim() orsimple_sim().

Usage

## S3 method for class 'bayes_ctd_array'print(  x = NULL,  measure = "power",  tab_type = "WX|YZ",  subj_per_arm_val = NULL,  a0_val = NULL,  effect_val = NULL,  rand_control_diff_val = NULL,  print_chg_warn = 1,  ...)

Arguments

Details

If the object of classbayes_ctd_array is created byhistoric_sim(),then the functionprint() allows the user to print user-specified 1- and 2-dimensional slices of the simulation results based on slicing code describedbelow. If the object of classbayes_ctd_array is created bysimple_sim(), a basic table of characteristic by sample size and effect is created.

If the object of classbayes_ctd_array is created bysimple_sim(), thenall four trial characteristics (subj_per_arm_val,a0_vals,effect_val, andrand_control_diff_val) can be ignored, as can theparameter defining what type of table to print,tab_type. A call toprint() will require the user to specify a measure (power, est, var, bias,or mse).

If the object of classbayes_ctd_array is created byhistoric_sim(),a call toprint() will require the user to specify a measure(power, est, var, bias, or mse) and may require the user to specify a table type.A table type,tab_type, will be required if 3 of the 4 trial characteristicsare equal to a vector of 2 or more values. The table type specificationuses the letters W, X, Y, and Z. The letter W represents the subject per armdimension. The letter X represents the a0 dimension. The letter Y representsthe effect dimension. The letter Z represents the control difference dimension.To define a slice of the 4-dimensional array, these letters are put into an AB|CDpattern. The two letters to the right of the vertical bar define which variablesare held constant. The two letters to the left of the vertical bar define whichvariables are going to show up in the rows (first letter) and in the columns (secondletter). For example if tab_type equalsWX|YZ, then effect and controldifferences will be held constant, while sample size will be represented by the rowsin the generated table and a0 values will be represented by the columns. The actualvalues that are printed in the tables depend on what measure is requested in theparametermeasure.

• tab_type='WX|YZ', Sample Size by a0

• tab_type='WY|XZ', Sample Size by Effect

• tab_type='WZ|XY', Sample Size by Control Differences

• tab_type='XY|WZ', a0 by Effect

• tab_type='XZ|WY', a0 by Control Differences

• tab_type='YZ|WX', Effect by Control Differences

• tab_type='ZX|WY', Control Differences by a0

• tab_type='XW|YZ', a0 by Sample Size

• tab_type='YW|XZ', Effect by Sample Size

• tab_type='YX|WZ', Effect by a0

• tab_type='ZW|XY', Control Differences by Sample Size

• tab_type='ZY|WX', Control Differences by Effect

It is very important to populate the values ofsubj_per_arm_val,a0_vals,effect_val, andrand_control_diff_val correctly giventhe value of tab_type, when the object of classbayes_ctd_array is created byhistoric_sim() and at least 3 of the four parameters have more than onevalue. On the other hand, if 2 or more of the four parameters have only one value,thensubj_per_arm_val,a0_vals,effect_val,rand_control_diff_val, as well astab_type can be ignored. If the lasttwo letters areYZ, theneffect_val andrand_control_diff_valmust be populated. If the last two letters areXZ, thena0_vals andrand_control_diff_val must be populated. If the last two letters areXY, thena0_vals andeffect_val must be populated. If the lasttwo letters areWZ, thensample_val andrand_control_diff_valmust be populated. If the last two letters areWY, thensample_size_valandeffect_val must be populated. If the last two letters areWX, thensample_size_val anda0_vals must be populated.

If the object of classbayes_ctd_array is created bysimple_sim(), theparameterstab_type,subj_per_arm_val,a0_vals,effect_val,andrand_control_diff_val are ignored.

Value

print() returns a two dimensional array of simulation results.

Examples

#Run a Weibull simulation, using simple_sim().#For meaningful results, trial_reps needs to be much larger than 2.weibull_test <- simple_sim(trial_reps = 2, outcome_type = "weibull",                           subj_per_arm = c(50, 100, 150, 200),                           effect_vals = c(0.6, 1, 1.4),                           control_parms = c(2.82487,3),                           time_vec = NULL, censor_value = NULL,                           alpha = 0.05, get_var = TRUE,                           get_bias = TRUE, get_mse = TRUE,                           seedval=123, quietly=TRUE)#Tabulate the simulation results for power.test_table <- print(x=weibull_test, measure="power",                    tab_type=NULL, subj_per_arm_val=NULL, a0_val=NULL,                    effect_val=NULL, rand_control_diff_val=NULL)print(test_table)#Tabulate the simulation results for estimates.print(x=weibull_test, measure="est")#Tabulate the simulation results for variance.print(x=weibull_test, measure="var")#Tabulate the simulation results for bias.print(x=weibull_test, measure="bias")#Tabulate the simulation results for mse.print(x=weibull_test, measure="mse")#Run another weibull simulation, using historic_sim().#Note: historic_sim() can take a while to run.#Generate a sample of historical data for use in example.set.seed(2250)SampleHistData <- genweibulldata(sample_size=60, scale1=2.82487,                                 hazard_ratio=0.6, common_shape=3,                                 censor_value=3)histdata <- subset(SampleHistData, subset=(treatment==0))histdata$id <- histdata$id+10000#For meaningful results, trial_reps needs to be larger than 100.weibull_test2 <- historic_sim(trial_reps = 100, outcome_type = "weibull",                              subj_per_arm = c(50, 100, 150, 200, 250),                              a0_vals = c(0, 0.33, 0.67, 1),                              effect_vals = c(0.6, 1, 1.4),                              rand_control_diff = c(0.8, 1, 1.2),                              hist_control_data = histdata, time_vec = NULL,                              censor_value = 3, alpha = 0.05, get_var = TRUE,                              get_bias = TRUE, get_mse = TRUE, seedval=123,                              quietly=TRUE)#Tabulate the simulation results for power.test_table <- print(x=weibull_test2, measure="power",                    tab_type="WX|YZ", effect_val=0.6,                    rand_control_diff_val=1.0)print(test_table)#Tabulate the simulation results for estimates.print(x=weibull_test2, measure="est", tab_type="WX|YZ",      effect_val=0.6, rand_control_diff_val=1.0)#Tabulate the simulation results for variance.print(x=weibull_test2, measure="var", tab_type="WX|YZ",      effect_val=0.6, rand_control_diff_val=1.0)#Tabulate the simulation results for bias.print(x=weibull_test2, measure="bias", tab_type="WX|YZ",      effect_val=0.6, rand_control_diff_val=1.0)#Tabulate the simulation results for mse.print(x=weibull_test2, measure="mse", tab_type="WX|YZ",      effect_val=0.6, rand_control_diff_val=1.0)#Run a Bernoulli simulation, using historic_sim().#Generate a sample of historical Bernoulli data for use in example.set.seed(2250)samplehistdata <- genbernoullidata(sample_size=60, prob1=0.6, odds_ratio=0.6)histdata <- subset(samplehistdata, subset=(treatment==0))histdata$id <- histdata$id+10000#For meaningful results, trial_reps needs to be larger than 100.bernoulli_test <- historic_sim(trial_reps = 100, outcome_type = "bernoulli",                              subj_per_arm = c(150),                              a0_vals = c(1.0),                              effect_vals = c(0.1,0.2,0.3,0.4,0.5,0.6,0.7,0.8,0.9,1.0),                              rand_control_diff = c(1.8),                              hist_control_data = histdata, time_vec = NULL,                              censor_value = 3, alpha = 0.05, get_var = TRUE,                              get_bias = TRUE, get_mse = TRUE, seedval=123,                              quietly=TRUE)test_table <- print(x=bernoulli_test, measure="power",                    tab_type=NULL, effect_val=NULL,                    subj_per_arm_val=NULL)print(test_table)#If only one or two of the subj_per_arm, a0_vals, effect_vals, or#rand_control_diff parameters have length greater than 1, then#only bayes_ctd_array and measure parameters are needed.#Tabulate the simulation results for estimates.print(x=bernoulli_test, measure="est")#Tabulate the simulation results for variance.print(x=bernoulli_test, measure="var")#Tabulate the simulation results for bias.print(x=bernoulli_test, measure="bias")#Tabulate the simulation results for mse.print(x=bernoulli_test, measure="mse")

Two Arm Bayesian Clinical Trial Simulation without Historical Data

Description

simple_sim() returns an S3 object of classbayes_ctd_array, whichwill contain simulation results for power, statistic estimation, bias, variance,and mse as requested by user.

Usage

simple_sim(  trial_reps = 100,  outcome_type = "weibull",  subj_per_arm = c(50, 100, 150, 200, 250),  effect_vals = c(0.6, 1, 1.4),  control_parms = NULL,  time_vec = NULL,  censor_value = NULL,  alpha = 0.05,  get_var = FALSE,  get_bias = FALSE,  get_mse = FALSE,  seedval = NULL,  quietly = TRUE)

Arguments

Details

The objectbayes_ctd_array has 6 elements: a list containing simulationresults (data), copies of the 4 function argumentssubj_per_arm,a0_vals,effect_vals, andrand_control_diff, and finallyaobjtype value indicating thatsimple_sim() was used. Each element ofdata is a four-dimensional array, where each dimension is determined by thelength of parameterssubj_per_arm,a0_vals,effect_vals, andrand_control_diff. The size ofdata depends on which results arerequested by the user. At a minimum, at least one ofsubj_per_arm,a0_vals,effect_vals, orrand_control_diff must contain atleast 2 values, while the other three must contain at least 1 value. Thedatalist will always contain two elements: an array of power results (power) andan array of estimation results (est). In addition topower andest,data may also contain elementsvar,bias, ormse, depending on the values ofget_var,get_bias, andget_mse. The values returned inest are in the form of hazard ratios,mean ratios, odds ratios, or mean differences depending on the value ofoutcome_type. For a Gaussian outcome, the estimation results aredifferences in group means (experimental group minus control group). For alogistic outcome, the estimation results are odds ratios (experimental group overcontrol group). For lognormal and Poisson outcomes, the estimation results are meanratios (experimental group over control group). For a piecewise exponential or aWeibull outcome, the estimation results are hazard ratios (experimental group overcontrol group). The values returned inbias,var, andmse areon the scale of the values returned inest.

The objectbayes_ctd_array has two primary methods,print() andplot(), for printing and plotting slices of the arrays contained inbayes_ctd_array$data.

As dimensions of the four dimensional array increases, the time required to completethe simulation will increase; however, it will be faster than a similar simulationbased on repeated calls to MCMC routines to analyze each simulated trial.

The meaning of the estimation results, and the test used to generate power results,depends on the outcome used. In all cases, power is based on a two-sided testinvolving a (1-alpha)100% credible interval, where the interval is used to determineif the null hypothesis should be rejected (null value outside of the interval) ornot rejected (null value inside the interval). For a Gaussian outcome, the 95%credible interval is an interval for the difference in group means(experimental group minus control group), and the test determines if 0 is in oroutside of the interval. For a Bernoulli outcome, the 95% credible intervalis an interval for the odds ratio (experimental group over control group),and the test determines if 1 is in or outside of the interval. For a lognormal ora Poisson outcome, the 95% credible interval is an interval for the mean ratio(experimental group over control group), and the test determines if 1 is in oroutside of the interval. Finally, for a piecewise exponential or a Weibull outcome,the 95% credible interval is an interval for the hazard ratio (experimental groupover control group), and the test determines if 1 is in or outside of the interval.

For a Gaussian outcome, thecontrol_parms values should be(mean, sd),where mean is the mean parameter for the control group used in a call tornorm(),and sd is the common sd parameter for both groups used in a call torlnorm().

For a Bernoulli outcome, thecontrol_parms values should be(prob), whereprob is the event probability for the control group used in a call torbinom().

For a lognormal outcome, thecontrol_parms values should be(meanlog, sdlog),where meanlog is the meanlog parameter for the control group used in a call torlnorm(), and sdlog is the common sdlog parameter for both groups used ina call torlnorm().

For a Poisson outcome, thecontrol_parms value should be(lambda), wherelambda is the lambda parameter for the control group used in a call torpois() andis equal to the mean of a Poisson distribution.

For a Weibull outcome, thecontrol_parms values should be(scale, shape),where scale is the scale parameter for the control group used in a call torweibull(), and shape is the common shape parameter for both groups used ina call torweibull().

For a piecewise exponential outcome, thecontrol_parms values should be a vectorof lambdas used in a call toeha::rpch(). Each element incontrol_parmsis a hazard for an interval defined by thetime_vec parameter.

Please refer to the examples for illustration of package use.

Value

simple_sim() returns an S3 object of classbayes_ctd_array.As noted in Details, an object of classbayes_ctd_array has 6 elements: alist containing simulation results (data), copies of the 4 functionargumentssubj_per_arm,a0_vals,effect_vals, andrand_control_diff, and finallyobjtype indicating thatsimple_sim()was used. See Details for a discussion about the contents ofdata. Results from the simulation contained in thebayes_ctd_arrayobject can be printed or plotted using theprint() andplot() methods. The results can also be accessed using basic listelement identification and array slicing. For example, to get the power resultsfrom a simulation, one could use the codebayes_ctd_array$data$power, wherebayes_ctd_array is replaced with the name of the variable containing thebayes_ctd_array object. Even though this is a 4-dimensional array, the powerresults only occupy a single 2-dimensional table. To print this 2-dimensional table,one would use the codebayes_ctd_array$data$power[,1,,1], wherebayes_ctd_array is replaced with the name of the variable containing thebayes_ctd_array object.

Examples

#Run a Weibull simulation, using simple_sim().#For meaningful results, trial_reps needs to be much larger than 2.weibull_test <- simple_sim(trial_reps = 2, outcome_type = "weibull",                           subj_per_arm = c(50, 100, 150, 200),                           effect_vals = c(0.6, 1, 1.4),                           control_parms = c(2.82487,3), time_vec = NULL,                           censor_value = NULL, alpha = 0.05,                           get_var = TRUE, get_bias = TRUE, get_mse = TRUE,                           seedval=123, quietly=TRUE)#Tabulate the simulation results for power.test_table <- print(x=weibull_test, measure="power",                    tab_type=NULL, subj_per_arm_val=NULL, a0_val=NULL,                    effect_val=NULL, rand_control_diff_val=NULL)print(test_table)#Create a plot of the power simulation results.plot(x=weibull_test, measure="power", tab_type=NULL,     smooth=FALSE, plot_out=TRUE)#Create a plot of the estimated hazard ratio simulation results.plot(x=weibull_test, measure="est", tab_type=NULL,     smooth=FALSE, plot_out=TRUE)#Create a plot of the hazard ratio variance simulation results.plot(x=weibull_test, measure="var", tab_type=NULL,     smooth=FALSE, plot_out=TRUE)#Create a plot of the hazard ratio bias simulation results.plot(x=weibull_test, measure="bias", tab_type=NULL,     smooth=FALSE, plot_out=TRUE)#Create a plot of the hazard ratio mse simulation results.plot(x=weibull_test, measure="mse", tab_type=NULL,     smooth=FALSE, plot_out=TRUE)