Multi-Armed Bandit Approaches to Pricing Experiments

---

Documentation for package ‘PricingBandits’ version 2.0.0

• DESCRIPTION file.
• User guides, package vignettes and other documentation.
• Package NEWS.

Help Pages

AggregateDataGP	Aggregate Data for Gaussian Process (GP) Algorithms
AggregateDataTS	Aggregate Data for Thompson Sampling (TS) Algorithm
AggregateDataUCB	GP Variants
BasisFunction	Basis Function
CovarianceFromKernel	Covariance Matrix from Kernel
GetDiagnostics	Get Experiment Diagnostics
GPTS	Gaussian Process Thompson Sampling (GPTS) Policy
GPTS_Mono	Gaussian Process Thompson Sampling Monotonic (GPTS_Mono) Policy
GPUCB	Gaussian Process Upper Confidence Bound (GPUCB) Policy
GPUCB_Mono	Gaussian Process Upper Confidence Bound Monotonic (GPUCB_Mono) Policy
JointCovFromKernel	Joint Covariance Matrix from Kernel
MABExperiment	Multi-Armed Bandit Experiment Framework
MakePosDefinitive	Positive Definite Covariance Matrix
NLML	Gaussian Process Regression
NoiseSample	Noise Sampling for Heteroscedastic Gaussian Processes
NonMonoPolicyEval	Evaluate Non-Monotonic Policies
OptimalHyperparameters	Optimal Hyperparameters
PolicyEvaluation	Evaluate Policies for Pricing Experiments
PosteriorPrediction	Posterior Prediction (Joint GP with Derivatives)
PricingBandit	Run a Pricing Bandit Experiment
RBFKernel	Kernel Functions
RBFKernel_01	RBF Kernel (Point to Derivative)
RBFKernel_11	RBF Kernel (Derivative to Derivative)
RBFKernel_All	Generalized RBF Kernel
ResetDiagnostics	Reset Experiment Diagnostics
TS	Thompson Sampling (TS) Policy
UCB	Bandit Policies for Pricing Experiments