Plotparms

DESCRIPTION

@@ -32,6 +32,7 @@ Imports:
     reshape2,
     lattice,
     grid,
+    gridExtra,
     tweedie,
     utils, 
     systemfit, 

NAMESPACE

@@ -1,5 +1,7 @@
 import(stats)
-importFrom(ggplot2, ggplot, geom_density, facet_wrap, aes, "scale_x_continuous")
+import(ggplot2)
+#importFrom(ggplot2, ggplot, geom_density, facet_wrap, aes, "scale_x_continuous")
+importFrom(gridExtra, marrangeGrob)
 importFrom(tweedie, rtweedie, "tweedie.profile")
 importFrom(statmod, tweedie)
 importFrom(systemfit, systemfit, systemfit.control)
@@ -20,7 +22,8 @@ importFrom(cplm, cpglm)
 importFrom(MASS, glm.nb)
 
 importClassesFrom(cplm, NullNum, NullList)
-importMethodsFrom(cplm, show, predict, vcov, residuals, resid, coef, fitted, plot, terms)
+importMethodsFrom(cplm, show, predict, vcov, residuals, resid, coef, fitted, 
+                  plot, terms)
 importMethodsFrom(Matrix, summary)
 
 export(MackChainLadder, MunichChainLadder, BootChainLadder,
@@ -59,7 +62,6 @@ S3method(print, ClarkLDF)
 S3method(print, ClarkCapeCod)
 S3method(print, ata)
 
-
 S3method(as.triangle, matrix)
 S3method(as.triangle, data.frame)
 S3method(as.data.frame, triangle)
@@ -89,8 +91,10 @@ S3method(mean, BootChainLadder)
 S3method(vcov, clark)
 
 exportMethods(predict, Mse, summary, show, coerce, 
-              "[", "$", "[[", "[<-", names, coef, vcov, residCov, residCor,
-              residuals, resid, rstandard, fitted, plot, cbind2, rbind2, dim)
+              "[", "$", "[[", "[<-", names, coef, 
+              vcov, residCov, residCor,
+              residuals, resid, rstandard, fitted, 
+              plot, cbind2, rbind2, dim)
 
 S3method(CDR, MackChainLadder)
 S3method(CDR, BootChainLadder)
@@ -101,4 +105,8 @@ export(tweedieReserve)
 S3method(print, tweedieReserve)
 S3method(summary, tweedieReserve)
 
+export(plotParms)
+S3method(plotParms, ChainLadder)
+S3method(plotParms, MackChainLadder)
+
 export(PaidIncurredChain)

R/ChainLadder.R

@@ -5,6 +5,8 @@
 
 chainladder <- function(Triangle, weights=1,
                         delta=1){
+
+  TriangleName <- deparse(substitute(Triangle))
 
     Triangle <- checkTriangle(Triangle)
     n <- dim(Triangle)[2]
@@ -27,7 +29,9 @@ chainladder <- function(Triangle, weights=1,
     } 
     myModel <- lapply(c(1:(n-1)), lmCL, Triangle)
 
-    output <- list(Models=myModel, Triangle=Triangle, delta=delta, weights=weights)
+    output <- list(Models=myModel, Triangle=Triangle, delta=delta, 
+                   weights=weights, TriangleName = TriangleName)
+    output[["call"]] <-  match.call(expand.dots = FALSE)
     class(output) <- c("ChainLadder", "TriangleModel", class(output))
     return(output)
 }

R/MackChainLadderFunctions.R

@@ -16,12 +16,22 @@ MackChainLadder <- function(
   tail.sigma=NULL,
   mse.method = "Mack")
 {
+  TriangleName <- deparse(substitute(Triangle))
     ## idea: have a list for tail factor
     ## tail=list(f=FALSE, f.se=NULL, sigma=NULL, F.se=NULL)
     ##
-    # 2013-02-25 Parameter risk recursive formula may have a third term per
+  tail.input <- tail
+  tail.se.input <- tail.se
+  tail.sigma.input <- tail.sigma
+  mse.method.input <- mse.method
+  # 2013-02-25 Parameter risk recursive formula may have a third term per
     #   Murphy and BBMW
     if (! mse.method %in% c("Mack", "Independence")) stop("mse.method must be 'Mack' or 'Independence'")
+    if (! est.sigma[1] %in% c("Mack", "log-linear")) {
+      if (!is.numeric(est.sigma)) 
+        stop("est.sigma must be 'Mack' or 'log-linear' or numeric")
+      }
+    if (!is.logical(tail) & !is.numeric(tail)) stop("tail must be logical or numeric")
 
     Triangle <- checkTriangle(Triangle)
     m <- dim(Triangle)[1]
@@ -118,6 +128,13 @@ MackChainLadder <- function(
     output[["Total.ProcessRisk"]] <- attr(Total.SE, "processrisk")
     output[["Total.ParameterRisk"]] <- attr(Total.SE, "paramrisk")
     output[["tail"]] <- tail
+    output[["TriangleName"]] <- TriangleName
+    output[["est.sigma"]] <- est.sigma
+    output[["tail.input"]] <- tail.input
+    output[["tail.se.input"]] <- tail.se.input
+    output[["tail.sigma.input"]] <- tail.sigma.input
+    output[["mse.method"]] <- mse.method
+    output[["mse.method.input"]] <- mse.method.input
     class(output) <- c("MackChainLadder", "TriangleModel", "list")
     return(output)
   }
@@ -347,15 +364,41 @@ tail.SE <- function(FullTriangle, StdErr, Total.SE, tail.factor, tail.se = NULL,
 
   if(is.null(tail.se)){
     .fse <- StdErr$f.se[start:(n-2)]
-    mse <- lm(log(.fse) ~ .dev)
-    tail.se <- exp(predict(mse, newdata=data.frame(.dev=tail.pos)))
+    ndx <- .fse > 0
+    numpos <- sum(ndx)
+    if (numpos) {
+      .devse <- .dev[ndx]
+      .fse <- .fse[ndx]
+      mse <- lm(log(.fse) ~ .devse)
+      if (numpos > 1) tail.se <- exp(predict(mse, newdata=data.frame(.devse=tail.pos)))
+      else {
+        warning("Only one column for estimating tail.se")
+        tail.se <- suppressWarnings(
+          exp(predict(mse, newdata=data.frame(.dev=tail.pos))))
+        }
+    }
+    else {
+      warning("No development variability in triangle. Setting tail.se = 0")
+      tail.se <- 0
+    }
   }
   StdErr$f.se <- c(StdErr$f.se, tail.se = tail.se)
 
   if(is.null(tail.sigma)){
     .sigma <- StdErr$sigma[start:(n-2)]
-    msig <- lm(log(.sigma) ~ .dev)
-    tail.sigma <- exp(predict(msig, newdata = data.frame(.dev = tail.pos)))
+    ndx <- .sigma > 0
+    numpos <- sum(ndx)
+    if (numpos) {
+      .devsigma <- .dev[ndx]
+      .sigma <- .sigma[ndx]
+      msig <- lm(log(.sigma) ~ .devsigma)
+      if (numpos > 1) tail.sigma <- exp(
+        predict(msig, newdata = data.frame(.devsigma = tail.pos)))
+      else {
+        warning("No development variability in triangle. Setting tail.sigma = 0")
+        tail.sigma <- 0
+      }
+    }
   }
   StdErr$sigma <- c(StdErr$sigma, tail.sigma = as.numeric(tail.sigma))