sufstat: Sufficient Statistics

In Boom: Bayesian Object Oriented Modeling

Sufficient Statistics

Description

Sufficient statistics for various models.

Usage

  RegressionSuf(X = NULL,
                y = NULL,
                xtx = crossprod(X),
                xty = crossprod(X, y),
                yty = sum(y^2),
                n = length(y),
                xbar = colMeans(X),
                ybar = mean(y))

  GaussianSuf(y)

Arguments

X	The predictor matrix for a regression problem.
y	The data, or the regression response variable.
xtx	The cross product of the design matrix. "X transpose X."
xty	The cross product of the design matrix with the response vector. "X transpose y."
yty	The sum of the squares of the response vector. "y transpose y."
n	The sample size.
xbar	A vector giving the average of each column in the predictor matrix.
ybar	The (scalar) mean of the response variable y.

Value

The returned value is a function containing the sufficient statistics for a regression model. Arguments are checked to ensure they have legal values. List names match the names expected by underlying C++ code.

Author(s)

Steven L. Scott [e-mail]

Examples

  X <- cbind(1, matrix(rnorm(3 * 100), ncol = 3))
  y <- rnorm(100)

  ## Sufficient statistics can be computed from raw data, if it is
  ## available.
  suf1 <- RegressionSuf(X, y)

  ## The individual components can also be computed elsewhere, and
  ## provided as arguments.  If n is very large, this can be a
  ## substantial coomputational savings.
  suf2 <- RegressionSuf(xtx = crossprod(X),
                        xty = crossprod(X, y),
                        yty = sum(y^2),
                        n = 100,
                        xbar = colMeans(X))

Boom documentation built on May 29, 2024, 5:08 a.m.