LossMatrix {LaplacesDemon}  R Documentation 
Loss Matrix
Description
A loss matrix is useful in Bayesian decision theory for selecting the Bayes action, the optimal Bayesian decision, when there are a discrete set of possible choices (actions) and a discrete set of possible outcomes (states of the world). The Bayes action is the action that minimizes expected loss, which is equivalent to maximizing expected utility.
Usage
LossMatrix(L, p.theta)
Arguments
L 
This required argument accepts a 
p.theta 
This required argument accepts a

Details
Bayesian inference is often tied to decision theory (Bernardo and Smith, 2000), and decision theory has long been considered the foundations of statistics (Savage, 1954).
Before using the LossMatrix
function, the user should have
already considered all possible actions (choices), states of the world
(outcomes unknown at the time of decisionmaking), chosen a loss
function L(\theta, \alpha)
, estimated loss, and
elicited prior probabilities p(\theta  x)
.
Although possible actions (choices) for the decisionmaker and possible states (outcomes) may be continuous or discrete, the loss matrix is used for discrete actions and states. An example of a continuous action may be that a decisionmaker has already decided to invest, and the remaining, current decision is how much to invest. Likewise, an example of continuous states of the world (outcomes) may be how much profit or loss may occur after a given continuous unit of time.
The coded example provided below is taken from Berger (1985, p. 67) and described here. The set of possible actions for a decisionmaker is to invest in bond ZZZ or alternatively in bond XXX, as it is called here. A realworld decision should include a mutually exhaustive list of actions, such as investing in neither, but perhaps the decisionmaker has already decided to invest and narrowed the options down to these two bonds.
The possible states of the world (outcomes unknown at the time of
decisionmaking) are considered to be two states: either the chosen
bond will not default or it will default. Here, the loss function is
a negative linear identity of money, and hence a loss in element
L[1,1]
of 500 is a profit of 500, while a loss in
L[2,1]
of 1,000 is a loss of 1,000.
The decisionmaker's dilemma is that bond ZZZ may return a higher profit than bond XXX, however there is an estimated 10% chance, the prior probability, that bond ZZZ will default and return a substantial loss. In contrast, bond XXX is considered to be a surething and return a steady but smaller profit. The Bayes action is to choose the first action and invest in bond ZZZ, because it minimizes expected loss, even though there is a chance of default.
A more realistic application of a loss matrix may be to replace the
pointestimates of loss with samples given uncertainty around the
estimated loss, and replace the pointestimates of the prior
probability of each state with samples given the uncertainty of the
probability of each state. The loss function used in the example is
intuitive, but a more popular monetary loss function may be
\log(E(W  R))
, the negative log of the
expectation of wealth, given the return. There are many alternative
loss functions.
Although isolated decisiontheoretic problems exist such as the
provided example, decision theory may also be applied to the results
of a probability model (such as from
IterativeQuadrature
, LaplaceApproximation
,
LaplacesDemon
, PMC
), or
VariationalBayes
, contingent on how
a decisionmaker is considering to use the information from the
model. The statistician may pass the results of a model to a client,
who then considers choosing possible actions, given this
information. The statistician should further assist the client with
considering actions, states of the world, then loss functions, and
finally eliciting the client's prior probabilities (such as with the
elicit
function).
When the outcome is finally observed, the information from this outcome may be used to refine the priors of the next such decision. In this way, Bayesian learning occurs.
Value
The LossMatrix
function returns a list with two components:
BayesAction 
This is a numeric scalar that indicates the action that minimizes expected loss. 
E.Loss 
This is a vector of expected losses, one for each action. 
Author(s)
Statisticat, LLC. software@bayesianinference.com
References
Berger, J.O. (1985). "Statistical Decision Theory and Bayesian Analysis", Second Edition. Springer: New York, NY.
Bernardo, J.M. and Smith, A.F.M. (2000). "Bayesian Theory". John Wiley \& Sons: West Sussex, England.
Savage, L.J. (1954). "The Foundations of Statistics". John Wiley \& Sons: West Sussex, England.
See Also
elicit
,
IterativeQuadrature
,
LaplaceApproximation
,
LaplacesDemon
,
PMC
, and
VariationalBayes
.
Examples
library(LaplacesDemon)
### Pointestimated loss and state probabilities
L < matrix(c(500,1000,300,300), 2, 2)
rownames(L) < c("s[1]: !Defaults","s[2]: Defaults")
colnames(L) < c("a[1]: Buy ZZZ", "a[2]: Buy XXX")
L
p.theta < matrix(c(0.9, 0.1, 1, 0), 2, 2)
Fit < LossMatrix(L, p.theta)
### Pointestimated loss and samples of state probabilities
L < matrix(c(500,1000,300,300), 2, 2)
rownames(L) < c("s[1]: Defaults","s[2]: !Defaults")
colnames(L) < c("a[1]: Buy ZZZ", "a[2]: Buy XXX")
L
p.theta < array(runif(4000), dim=c(2,2,1000)) #Random probabilities,
#just for a quick example. And, since they must sum to one:
for (i in 1:1000) {
p.theta[,,i] < p.theta[,,i] / matrix(colSums(p.theta[,,i]),
dim(p.theta)[1], dim(p.theta)[2], byrow=TRUE)}
Fit < LossMatrix(L, p.theta)
Fit
### Pointestimates of loss may be replaced with samples as well.