change "bias" to "mu"

R/bmm_model_IMM.R

@@ -166,11 +166,11 @@ configure_model.IMMabc <- function(model, data, formula) {
   pform_names <- names(formula)
   pform <- formula
 
-  # add fixed intercept for bias if no formula was included
-  if (!"bias" %in% pform_names) {
-    bias_form <- bias ~ 1
-    pform <- c(pform, bias_form)
-    names(pform) <- c(pform_names,"bias")
+  # add fixed intercept for mu if no formula was included
+  if (!"mu" %in% pform_names) {
+    mu_form <- mu ~ 1
+    pform <- c(pform, mu_form)
+    names(pform) <- c(pform_names,"mu")
   }
 
   # names for parameters
@@ -181,7 +181,7 @@ configure_model.IMMabc <- function(model, data, formula) {
   mu_unif <- paste0('mu', max_setsize + 1)
 
   # construct formula
-  formula <- brms::bf(paste0(respErr,"~ bias"), nl = T)
+  formula <- brms::bf(paste0(respErr,"~ mu"), nl = T)
 
   # add parameter formulas to model formula
   for (i in 1:length(pform)) {
@@ -213,7 +213,7 @@ configure_model.IMMabc <- function(model, data, formula) {
 
   # define prior
   prior <- # fix mean of the first von Mises to zero
-    brms::prior_("constant(0)", nlpar = "bias") +
+    brms::prior_("constant(0)", nlpar = "mu") +
     # fix mean of the guessing distribution to zero
     brms::prior_("constant(0)", class = "Intercept", dpar = mu_unif) +
     # fix kappa of the second von Mises to (alomst) zero
@@ -247,11 +247,11 @@ configure_model.IMMbsc <- function(model, data, formula) {
   pform_names <- names(formula)
   pform <- formula
 
-  # add fixed intercept for bias if no formula was included
-  if (!"bias" %in% pform_names) {
-    bias_form <- bias ~ 1
-    pform <- c(pform, bias_form)
-    names(pform) <- c(pform_names,"bias")
+  # add fixed intercept for mu if no formula was included
+  if (!"mu" %in% pform_names) {
+    mu_form <- mu ~ 1
+    pform <- c(pform, mu_form)
+    names(pform) <- c(pform_names,"mu")
   }
 
   # names for parameters
@@ -262,7 +262,7 @@ configure_model.IMMbsc <- function(model, data, formula) {
   mu_unif <- paste0('mu', max_setsize + 1)
 
   # construct formula
-  formula <- brms::bf(paste0(respErr,"~ bias"), nl = T)
+  formula <- brms::bf(paste0(respErr,"~ mu"), nl = T)
 
   # add parameter formulas to model formula
   for (i in 1:length(pform)) {
@@ -295,7 +295,7 @@ configure_model.IMMbsc <- function(model, data, formula) {
 
   # define prior
   prior <- # fix mean of the first von Mises to zero
-    brms::prior_("constant(0)", nlpar = "bias") +
+    brms::prior_("constant(0)", nlpar = "mu") +
     # fix mean of the guessing distribution to zero
     brms::prior_("constant(0)", class = "Intercept", dpar = mu_unif) +
     # fix kappa of the second von Mises to (alomst) zero
@@ -329,11 +329,11 @@ configure_model.IMMfull <- function(model, data, formula) {
   pform_names <- names(formula)
   pform <- formula
 
-  # add fixed intercept for bias if no formula was included
-  if (!"bias" %in% pform_names) {
-    bias_form <- bias ~ 1
-    pform <- c(pform, bias_form)
-    names(pform) <- c(pform_names,"bias")
+  # add fixed intercept for mu if no formula was included
+  if (!"mu" %in% pform_names) {
+    mu_form <- mu ~ 1
+    pform <- c(pform, mu_form)
+    names(pform) <- c(pform_names,"mu")
   }
 
   # names for parameters
@@ -344,7 +344,7 @@ configure_model.IMMfull <- function(model, data, formula) {
   mu_unif <- paste0('mu', max_setsize + 1)
 
   # construct formula
-  formula <- brms::bf(paste0(respErr,"~ bias"), nl = T)
+  formula <- brms::bf(paste0(respErr,"~ mu"), nl = T)
 
   # add parameter formulas to model formula
   for (i in 1:length(pform)) {
@@ -377,7 +377,7 @@ configure_model.IMMfull <- function(model, data, formula) {
 
   # define prior
   prior <- # fix mean of the first von Mises to zero
-    brms::prior_("constant(0)", nlpar = "bias") +
+    brms::prior_("constant(0)", nlpar = "mu") +
     # fix mean of the guessing distribution to zero
     brms::prior_("constant(0)", class = "Intercept", dpar = mu_unif) +
     # fix kappa of the second von Mises to (alomst) zero

R/bmm_model_mixture2p.R

@@ -65,14 +65,14 @@ configure_model.mixture2p <- function(model, data, formula) {
   pform_names <- names(formula)
   pform <- formula
 
-  if (!"bias" %in% pform_names) {
-    bias_form <- bias ~ 1
-    pform <- c(pform, bias_form)
-    names(pform) <- c(pform_names,"bias")
+  if (!"mu" %in% pform_names) {
+    mu_form <- mu ~ 1
+    pform <- c(pform, mu_form)
+    names(pform) <- c(pform_names,"mu")
   }
 
   # specify the formula for the mixture model
-  formula <- brms::bf(paste0(respErr,"~ bias"), nl = T)
+  formula <- brms::bf(paste0(respErr,"~ mu"), nl = T)
 
   # add parameter formulas to model formula
   for (i in 1:length(pform)) {
@@ -95,7 +95,7 @@ configure_model.mixture2p <- function(model, data, formula) {
 
   # set priors for the estimated parameters
   prior <- # fix mean of the first von Mises to zero
-    brms::prior_("constant(0)", nlpar = "bias") +
+    brms::prior_("constant(0)", nlpar = "mu") +
     # fix mean of the second von Mises to zero
     brms::prior_("constant(0)", class = "Intercept", dpar = "mu2") +
     # fix kappa of the second von Mises to (alomst) zero

R/bmm_model_mixture3p.R

@@ -90,11 +90,11 @@ configure_model.mixture3p <- function(model, data, formula) {
   pform_names <- names(formula)
   pform <- formula
 
-  # add fixed intercept for bias if no formula was included
-  if (!"bias" %in% pform_names) {
-    bias_form <- bias ~ 1
-    pform <- c(pform, bias_form)
-    names(pform) <- c(pform_names,"bias")
+  # add fixed intercept for mu if no formula was included
+  if (!"mu" %in% pform_names) {
+    mu_form <- mu ~ 1
+    pform <- c(pform, mu_form)
+    names(pform) <- c(pform_names,"mu")
   }
 
   # names for parameters
@@ -105,7 +105,7 @@ configure_model.mixture3p <- function(model, data, formula) {
   mu_unif <- paste0('mu', max_setsize + 1)
 
   # construct formula
-  formula <- brms::bf(paste0(respErr,"~ bias"), nl = T)
+  formula <- brms::bf(paste0(respErr,"~ mu"), nl = T)
 
   # add parameter formulas to model formula
   for (i in 1:length(pform)) {
@@ -137,7 +137,7 @@ configure_model.mixture3p <- function(model, data, formula) {
 
   # define prior
   prior <-
-    brms::prior_("constant(0)", nlpar = "bias") +
+    brms::prior_("constant(0)", nlpar = "mu") +
     brms::prior_("constant(0)", class = "Intercept", dpar = mu_unif) +
     brms::prior_("constant(-100)", class = "Intercept", dpar = kappa_unif) +
     brms::prior_("normal(2, 1)", class = "b", nlpar = "kappa") +

R/bmm_model_sdmSimple.R

@@ -107,14 +107,14 @@ configure_model.sdmSimple <- function(model, data, formula) {
     pform_names <- names(formula)
     pform <- formula
 
-    if (!"bias" %in% pform_names) {
-      bias_form <- bias ~ 1
-      pform <- c(pform, bias_form)
-      names(pform) <- c(pform_names,"bias")
+    if (!"mu" %in% pform_names) {
+      mu_form <- mu ~ 1
+      pform <- c(pform, mu_form)
+      names(pform) <- c(pform_names,"mu")
     }
 
     # specify the formula for the mixture model
-    formula <- brms::bf(paste0(respErr,"~ bias"), nl = T)
+    formula <- brms::bf(paste0(respErr,"~ mu"), nl = T)
 
     # add parameter formulas to model formula
     for (i in 1:length(pform)) {
@@ -130,7 +130,7 @@ configure_model.sdmSimple <- function(model, data, formula) {
    # TODO: add a proper prior
    prior <-
      # fix mu to 0 (when I change mu to be the center, not c)
-     brms::prior_("constant(0)", nlpar = "bias")
+     brms::prior_("constant(0)", class = "Intercept", dpar = "mu")
 
    # set initial values to be sampled between [-1,1] to avoid extreme SDs that
    # can cause the sampler to fail // TODO: test extensively if this works in