From 9a96391589fef9f80f9c6c4881cc48a509cb75f2 Mon Sep 17 00:00:00 2001
From: Johannes Ranke <jranke@uni-bremen.de>
Date: Tue, 4 Jun 2019 11:15:52 +0200
Subject: Algorithms direct, two-, three-, fourstep, IRLS

All of them are working now and allow for comparison
Based on SFO, DFOP and HS fits to twelve test datasets, only
the combination of direct and threestep is needed to find
the lowest AIC
---
 R/mkinfit.R | 50 +++++++++++++++++++++++++++++++++++++++++---------
 1 file changed, 41 insertions(+), 9 deletions(-)

(limited to 'R/mkinfit.R')

diff --git a/R/mkinfit.R b/R/mkinfit.R
index 224dc7bd..e0a0e525 100644
--- a/R/mkinfit.R
+++ b/R/mkinfit.R
@@ -35,6 +35,7 @@ mkinfit <- function(mkinmod, observed,
   atol = 1e-8, rtol = 1e-10, n.outtimes = 100,
   error_model = c("const", "obs", "tc"),
   error_model_algorithm = c("direct", "twostep", "threestep", "fourstep", "IRLS"),
+  reweight.tol = 1e-8, reweight.max.iter = 10,
   trace_parms = FALSE,
   ...)
 {
@@ -285,7 +286,7 @@ mkinfit <- function(mkinmod, observed,
     # Trace parameter values if requested and if we are actually optimising
     if(trace_parms & update_data) cat(P, "\n")
 
-    if (fixed_degparms[1] != FALSE) {
+    if (is.numeric(fixed_degparms)) {
       degparms <- fixed_degparms
       errparms <- P # This version of errparms is local to the function
       degparms_fixed = TRUE
@@ -293,7 +294,7 @@ mkinfit <- function(mkinmod, observed,
       degparms_fixed = FALSE
     }
 
-    if (fixed_errparms[1] != FALSE) {
+    if (is.numeric(fixed_errparms)) {
       degparms <- P
       errparms <- fixed_errparms # Local to the function
       errparms_fixed = TRUE
@@ -364,7 +365,8 @@ mkinfit <- function(mkinmod, observed,
         sum(dnorm(x = value.observed, mean = value.predicted, sd = std, log = TRUE)))
     }
 
-    # We update the current likelihood and data during the optimisation, not during hessian calculations
+    # We update the current likelihood and data during the optimisation, not
+    # during hessian calculations
     if (update_data) {
 
       assign("out_predicted", out_long, inherits = TRUE)
@@ -434,11 +436,10 @@ mkinfit <- function(mkinmod, observed,
   # Show parameter names if tracing is requested
   if(trace_parms) cat(names_optim, "\n")
 
+  # browser()
+
   # Do the fit and take the time until the hessians are calculated
   fit_time <- system.time({
-    # In the inital run, we assume a constant standard deviation and do
-    # not optimise it, as this is unnecessary and leads to instability of the
-    # fit (most obvious when fitting the IORE model)
     degparms <- c(state.ini.optim, transparms.optim)
 
     if (err_mod == "const" | error_model_algorithm != "direct") {
@@ -463,7 +464,7 @@ mkinfit <- function(mkinmod, observed,
       fit <- nlminb(errparms, nlogLik, control = control,
         lower = lower[names(errparms)],
         upper = upper[names(errparms)],
-        degparms_fixed = degparms, ...)
+        fixed_degparms = degparms, ...)
       errparms <- fit$par
     }
     if (error_model_algorithm == "fourstep") {
@@ -471,10 +472,11 @@ mkinfit <- function(mkinmod, observed,
       fit <- nlminb(degparms, nlogLik, control = control,
         lower = lower[names(degparms)],
         upper = upper[names(degparms)],
-        errparms_fixed = errparms, ...)
+        fixed_errparms = errparms, ...)
       degparms <- fit$par
     }
-    if (error_model_algorithm %in% c("direct", "twostep", "threestep", "fourstep")) {
+    if (error_model_algorithm %in% c("direct", "twostep", "threestep", "fourstep") &
+        err_mod != "const") {
       if (!quiet) message("Optimising the complete model")
       parms.start <- c(degparms, errparms)
       fit <- nlminb(parms.start, nlogLik,
@@ -483,6 +485,36 @@ mkinfit <- function(mkinmod, observed,
         control = control, ...)
       fit$logLik <- - nlogLik.current
     }
+    if (err_mod != "const" & error_model_algorithm == "IRLS") {
+      reweight.diff <- 1
+      n.iter <- 0
+      errparms_last <- errparms
+
+      while (reweight.diff > reweight.tol &
+             n.iter < reweight.max.iter) {
+
+        if (!quiet) message("Optimising the error model")
+        fit <- nlminb(errparms, nlogLik, control = control,
+          lower = lower[names(errparms)],
+          upper = upper[names(errparms)],
+          fixed_degparms = degparms, ...)
+        errparms <- fit$par
+
+        if (!quiet) message("Optimising the degradation model")
+        fit <- nlminb(degparms, nlogLik, control = control,
+          lower = lower[names(degparms)],
+          upper = upper[names(degparms)],
+          fixed_errparms = errparms, ...)
+        degparms <- fit$par
+
+        reweight.diff <- dist(rbind(errparms, errparms_last))
+        errparms_last <- errparms
+
+        fit$par <- c(fit$par, errparms)
+        nlogLik.current <- nlogLik(c(degparms, errparms), OLS = FALSE)
+        fit$logLik <- - nlogLik.current
+      }
+    }
 
     # We include the error model in the parameter uncertainty analysis, also
     # for constant variance, to get a confidence interval for it
-- 
cgit v1.2.1