#' o$f( o$linArgs(par0) )
#' o$f( o$linArgs(par1) )
#' @export
-optimParams <- function(X, Y, K, link=c("logit","probit"))
+optimParams <- function(X, Y, K, link=c("logit","probit"), M=NULL)
{
# Check arguments
if (!is.matrix(X) || any(is.na(X)))
if (!is.numeric(K) || K!=floor(K) || K < 2)
stop("K: integer >= 2")
+ if (is.null(M))
+ {
+ # Precompute empirical moments
+ Mtmp <- computeMoments(X, Y)
+ M1 <- as.double(Mtmp[[1]])
+ M2 <- as.double(Mtmp[[2]])
+ M3 <- as.double(Mtmp[[3]])
+ M <- c(M1, M2, M3)
+ }
+ else
+ M <- c(M[[1]], M[[2]], M[[3]])
+
# Build and return optimization algorithm object
methods::new("OptimParams", "li"=link, "X"=X,
- "Y"=as.integer(Y), "K"=as.integer(K))
+ "Y"=as.integer(Y), "K"=as.integer(K), "Mhat"=as.double(M))
}
#' Encapsulated optimization for p (proportions), β and b (regression parameters)
"Check args and initialize K, d, W"
callSuper(...)
- if (!hasArg("X") || !hasArg("Y") || !hasArg("K") || !hasArg("li"))
+ if (!hasArg("X") || !hasArg("Y") || !hasArg("K")
+ || !hasArg("li") || !hasArg("Mhat"))
+ {
stop("Missing arguments")
-
- # Precompute empirical moments
- M <- computeMoments(X, Y)
- M1 <- as.double(M[[1]])
- M2 <- as.double(M[[2]])
- M3 <- as.double(M[[3]])
- Mhat <<- c(M1, M2, M3)
+ }
n <<- nrow(X)
- d <<- length(M1)
- W <<- diag(d+d^2+d^3) #initialize at W = Identity
+ d <<- ncol(X)
+ # W will be initialized when calling run()
},
expArgs = function(v)
list(
# p: dimension K-1, need to be completed
"p" = c(v[1:(K-1)], 1-sum(v[1:(K-1)])),
- "β" = matrix(v[K:(K+d*K-1)], ncol=K),
+ "β" = t(matrix(v[K:(K+d*K-1)], ncol=d)),
"b" = v[(K+d*K):(K+(d+1)*K-1)])
},
{
"Linearize vectors+matrices from list L into a vector"
- c(L$p[1:(K-1)], as.double(L$β), L$b)
+ # β linearized row by row, to match derivatives order
+ c(L$p[1:(K-1)], as.double(t(L$β)), L$b)
},
- #TODO: compare with R version?
- #D <- diag(d) #matrix of ej vectors
- #Y * X
- #Y * ( t( apply(X, 1, function(row) row %o% row) ) - Reduce('+', lapply(1:d, function(j) as.double(D[j,] %o% D[j,])), rep(0, d*d)))
- #Y * ( t( apply(X, 1, function(row) row %o% row %*% row) ) - Reduce('+', lapply(1:d, function(j) ), rep(0, d*d*d)))
computeW = function(θ)
{
- #require(MASS)
+ require(MASS)
dd <- d + d^2 + d^3
M <- Moments(θ)
Omega <- matrix( .C("Compute_Omega",
- X=as.double(X), Y=as.double(Y), M=as.double(M),
+ X=as.double(X), Y=as.integer(Y), M=as.double(M),
pn=as.integer(n), pd=as.integer(d),
W=as.double(W), PACKAGE="morpheus")$W, nrow=dd, ncol=dd )
- W <<- MASS::ginv(Omega, tol=1e-4)
- NULL #avoid returning W
+ MASS::ginv(Omega)
},
Moments = function(θ)
"Gradient of f, dimension (K-1) + d*K + K = (d+2)*K - 1"
L <- expArgs(θ)
- -2 * t(grad_M(L)) %*% W %*% as.matrix((Mhat - Moments(L)))
+ -2 * t(grad_M(L)) %*% W %*% as.matrix(Mhat - Moments(L))
},
grad_M = function(θ)
stop("θ0: list")
if (is.null(θ0$β))
stop("At least θ0$β must be provided")
- if (!is.matrix(θ0$β) || any(is.na(θ0$β)) || ncol(θ0$β) != K)
- stop("θ0$β: matrix, no NA, ncol == K")
+ if (!is.matrix(θ0$β) || any(is.na(θ0$β))
+ || nrow(θ0$β) != d || ncol(θ0$β) != K)
+ {
+ stop("θ0$β: matrix, no NA, nrow = d, ncol = K")
+ }
if (is.null(θ0$p))
θ0$p = rep(1/K, K-1)
- else if (length(θ0$p) != K-1 || sum(θ0$p) > 1)
- stop("θ0$p should contain positive integers and sum to < 1")
- # Next test = heuristic to detect missing b (when matrix is called "beta")
- if (is.null(θ0$b) || all(θ0$b == θ0$β))
+ else if (!is.numeric(θ0$p) || length(θ0$p) != K-1
+ || any(is.na(θ0$p)) || sum(θ0$p) > 1)
+ {
+ stop("θ0$p: length K-1, no NA, positive integers, sum to <= 1")
+ }
+ if (is.null(θ0$b))
θ0$b = rep(0, K)
- else if (any(is.na(θ0$b)))
- stop("θ0$b cannot have missing values")
- # TODO: stopping condition? N iterations? Delta <= epsilon ?
- for (loop in 1:10)
+ else if (!is.numeric(θ0$b) || length(θ0$b) != K || any(is.na(θ0$b)))
+ stop("θ0$b: length K, no NA")
+
+ # (Re)Set W to identity, to allow several run from the same object
+ W <<- diag(d+d^2+d^3)
+
+ loopMax <- 2 #TODO: loopMax = 3 ? Seems not improving...
+ x_init <- linArgs(θ0)
+ for (loop in 1:loopMax)
{
- op_res = constrOptim( linArgs(θ0), .self$f, .self$grad_f,
+ op_res = constrOptim( x_init, .self$f, .self$grad_f,
ui=cbind(
rbind( rep(-1,K-1), diag(K-1) ),
matrix(0, nrow=K, ncol=(d+1)*K) ),
ci=c(-1,rep(0,K-1)) )
-
- computeW(expArgs(op_res$par))
- # debug:
- #print(W)
- print(op_res$value)
- print(expArgs(op_res$par))
+ if (loop < loopMax) #avoid computing an extra W
+ W <<- computeW(expArgs(op_res$par))
+ x_init <- op_res$par
+ #print(op_res$value) #debug
+ #print(expArgs(op_res$par)) #debug
}
expArgs(op_res$par)