Description Usage Arguments Value Examples
This function computes a "Maximum Constrained Likelihood Estimator" by using optim() to optimize a constrained log-likelihood function. The constraint is a function that is applied to the log-likelihood of each observation which, in the ideal case, prevents the log-likelihood from being too small and hence the observation too influential on the estimator.
1 2 | MCLE(data, dist, initial, returnOptim = FALSE,
constraintFunc = constraintET, constraintFuncDeriv = constraintDerivET)
|
data |
A matrix of data. If univariate, data should have 1 column. If multivariate, data should have one column for each dimension. |
dist |
A distribution object (essentially a list with functions defining the distribution, but see the distribution class defined in this package). See getNormalDistribution and the functions it loads for examples of how to define this distribution object and the corresponding functions. |
initial |
Initial estimate for the parameters. This should be a list of parameters such that it could be passed to the paramList2Vec function in the distribution object. |
returnOptim |
Logical. If FALSE (default) the function returns the numeric estimates of the parameters, converted back into a list object. If TRUE, the function returns a list with the results of the call to optim (named "optim" in the output) as well as the parameters (named "solution") |
constraintFunc |
A function taking a numeric vector of likelihood values and returning the adjusted likelihood values. |
constraintFuncDeriv |
A function providing the derivative of the constraint function. This function should also accept a numeric vector and return a numeric vector of derivative values. |
See the returnOptim argument description.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 | ## Not run:
data = matrix(rnorm(200), nrow = 100)
mst = list(dev = MCLE:::devMST, grad = MCLE:::gradDevMST,
paramList2Vec = MCLE:::paramList2VecMST,
paramVec2List = MCLE:::paramVec2ListMST)
initial = list(xi = c(0, 0), Omega = diag(c(1, 1)),
alpha = c(0, 0), nu = 100)
MCLE(data, dist = mst, initial)
mt = list(dev = MCLE:::devMT, grad = MCLE:::gradDevMT,
paramList2Vec = MCLE:::paramList2VecMT,
paramVec2List = MCLE:::paramVec2ListMT)
initial = list(xi = c(0, 0), Omega = diag(c(1, 1)), nu = 100)
MCLE(data, dist = mt, initial)
msn = list(dev = MCLE:::devMSN, grad = MCLE:::gradDevMSN,
paramList2Vec = MCLE:::paramList2VecMSN,
paramVec2List = MCLE:::paramVec2ListMSN)
initial = list(xi = c(0, 0), Omega = diag(c(1, 1)), alpha = c(0, 0))
MCLE(data, dist = msn, initial)
mn = list(dev = MCLE:::devMN, grad = MCLE:::gradDevMN,
paramList2Vec = MCLE:::paramList2VecMN,
paramVec2List = MCLE:::paramVec2ListMN)
initial = list(xi = c(0, 0), Omega = diag(c(1, 1)), alpha = c(0, 0))
MCLE(data, dist = msn, initial)
data = rpois(30, lambda = 4.7)
data = c(data, 100)
dist = list(dev = MCLE:::devPsn, grad = MCLE:::gradDevPsn,
paramList2Vec = MCLE:::paramList2VecPsn,
paramVec2List = MCLE:::paramVec2ListPsn)
initial = list(lambda = 1)
MCLE(data, dist = dist, initial)
mean(data)
## End(Not run)
|
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.