BayesPPD (version 1.1.3)

normalizing.constant: Function for approximating the normalizing constant for generalized linear models with random a0

Description

This function returns a vector of coefficients that defines a function \(f(a_0)\) that approximates the normalizing constant for generalized linear models with random \(a_0\). The user should input the values returned to glm.random.a0 or power.glm.random.a0.

Usage

normalizing.constant(
  grid,
  historical,
  data.type,
  data.link,
  prior.beta.var = rep(10, 50),
  lower.limits = rep(-100, 50),
  upper.limits = rep(100, 50),
  slice.widths = rep(1, 50),
  nMC = 10000,
  nBI = 250
)

Value

Vector of coefficients for \(a_0\) that defines a function \(f(a_0)\) that approximates the normalizing constant, necessary for functions glm.random.a0 and power.glm.random.a0. The length of the vector is equal to 1+K*L where K is the number of historical datasets and L is the degree of the polynomial regression determined by the algorithm.

Arguments

grid

Matrix of potential values for \(a_0\), where the number of columns should equal the number of historial datasets. Note that the algorithm may fail if some grid values are close to zero. See Details below.

historical

List of historical dataset(s). East historical dataset is stored in a list which constains two named elements: y0 and x0.

• y0 is a vector of responses.

• x0 is a matrix of covariates.

For binomial data, an additional element n0 is required.

• n0 is vector of integers specifying the number of subjects who have a particular value of the covariate vector.

data.type

Character string specifying the type of response. The options are "Bernoulli", "Binomial", "Poisson" and "Exponential".

data.link

Character string specifying the link function. The options are "Logistic", "Probit", "Log", "Identity-Positive", "Identity-Probability" and "Complementary Log-Log". Does not apply if data.type is "Normal".

prior.beta.var

Vector of variances of the independent normal initial priors on \(\beta\) with mean zero. The length of the vector should be equal to the length of \(\beta\). The default variance is 10.

lower.limits

Vector of lower limits for parameters to be used by the slice sampler. The length of the vector should be equal to the total number of parameters, i.e. P+1 where P is the number of covariates. The default is -100 for all parameters (may not be appropriate for all situations). Does not apply if data.type is "Normal".

upper.limits

Vector of upper limits for parameters to be used by the slice sampler. The length of the vector should be equal to the total number of parameters, i.e. P+1 where P is the number of covariates. The default is 100 for all parameters (may not be appropriate for all situations). Does not apply if data.type is "Normal".

slice.widths

Vector of initial slice widths for parameters to be used by the slice sampler. The length of the vector should be equal to the total number of parameters, i.e. P+1 where P is the number of covariates. The default is 1 for all parameter (may not be appropriate for all situations). Does not apply if data.type is "Normal".

nMC

Number of iterations (excluding burn-in samples) for the slice sampler or Gibbs sampler. The default is 10,000.

nBI

Number of burn-in samples for the slice sampler or Gibbs sampler. The default is 250.

Details

This function performs the following steps:

1. Suppose there are K historical datasets. The user inputs a grid of M rows and K columns of potential values for \(a_0\). For example, one can choose the vector v = c(0.1, 0.25, 0.5, 0.75, 1) and use expand.grid(a0_1=v, a0_2=v, a0_3=v) when \(K=3\) to get a grid with \(M=5^3=125\) rows and 3 columns. If there are more than three historical datasets, the dimension of v can be reduced to limit the size of the grid. A large grid will increase runtime.

2. For each row of \(a_0\) values in the grid, obtain \(M\) samples for \(\beta\) from the power prior associated with the current values of \(a_0\) using the slice sampler.

3. For each of the M sets of posterior samples, execute the PWK algorithm (Wang et al., 2018) to estimate the log of normalizing constant \(d_1,...,d_M\) for the normalized power prior.

4. At this point, one has a dataset with outcomes \(d_1,...,d_M\) and predictors corresponding to the rows of the \(a_0\) grid matrix. A polynomial regression is applied to estimate a function \(d=f(a0)\). The degree of the polynomial regression is determined by the algorithm to ensure \(R^2 > 0.99\).

5. The vector of coefficients from the polynomial regression model is returned by the function, which the user must input into glm.random.a0 or power.glm.random.a0.

When a row of the grid contains elements that are close to zero, the resulting power prior will be flat and estimates of normalizing constants may be inaccurate. Therefore, it is recommended that grid values should be at least 0.05.

If one encounters the error message "some coefficients are not defined because of singularities", it could be due to the following factors: number of grid rows too large or too small, insufficient sample size of the historical data, insufficient number of iterations for the slice sampler, or near-zero grid values.

Note that due to computational intensity, the normalizing.constant function has not been evaluated for accuracy for high dimensional \(\beta\) (e.g., dimension > 10) or high dimensional \(a_0\) (e.g., dimension > 5).

References

Wang, Yu-Bo; Chen, Ming-Hui; Kuo, Lynn; Lewis, Paul O. A New Monte Carlo Method for Estimating Marginal Likelihoods. Bayesian Anal. 13 (2018), no. 2, 311--333.

See Also

glm.random.a0 and power.glm.random.a0

Examples

data.type <- "Bernoulli"
data.link <- "Logistic"
data.size <- 50

# Simulate two historical datasets
p <- 1
set.seed(111)
x1 <- matrix(rnorm(p*data.size),ncol=p,nrow=data.size)
set.seed(222)
x2 <- matrix(rnorm(p*data.size),ncol=p,nrow=data.size)
beta <- c(1,2)
mean1 <- exp(x1*beta)/(1+exp(x1*beta))
mean2 <- exp(x2*beta)/(1+exp(x2*beta))
historical <- list(list(y0=rbinom(data.size,size=1,prob=mean1),x0=x1),
                   list(y0=rbinom(data.size, size=1, prob=mean2),x0=x2))

# Create grid of possible values of a0 with two columns corresponding to a0_1 and a0_2
g <- c(0.1, 0.25, 0.5, 0.75, 1)
grid <- expand.grid(a0_1=g, a0_2=g)

nMC <- 100 # nMC should be larger in practice
nBI <- 50
result <- normalizing.constant(grid=grid, historical=historical,
                               data.type=data.type, data.link=data.link,
                               nMC=nMC, nBI=nBI)