Run cannonicalizer on all models

WardBrian · WardBrian · commit 6c30f0ea8e76 · 2023-03-14T14:56:02.000-04:00
diff --git a/ARM/Ch.17/radon_multi_varying_coef_17.1.stan b/ARM/Ch.17/radon_multi_varying_coef_17.1.stan
@@ -22,35 +22,35 @@ model {
   vector[K] mu;
   matrix[J, K] B_raw;
   vector[K] sigma_b;
-  
+
   mu_raw ~ normal(0, 100);
   xi ~ uniform(0, 100);
-  
+
   mu = xi .* mu_raw;
-  
+
   Tau_b_raw ~ wishart(K + 1, W);
   Sigma_b_raw = inverse(Tau_b_raw);
-  
+
   for (j in 1 : J) {
     B_raw_temp ~ multi_normal(mu_raw, Sigma_b_raw);
     for (k in 1 : K) {
       B_raw[j, k] = B_raw_temp[k];
       B[j, k] = xi[k] * B_raw[j, k];
     }
   }
-  
+
   for (k in 1 : K) {
     for (k_prime in 1 : K) {
       rho_b[k, k_prime] = Sigma_b_raw[k, k_prime]
                           / sqrt(Sigma_b_raw[k, k]
                                  * Sigma_b_raw[k_prime, k_prime]);
     }
-    sigma_b[k] = fabs(xi[k]) * sqrt(Sigma_b_raw[k, k]);
+    sigma_b[k] = abs(xi[k]) * sqrt(Sigma_b_raw[k, k]);
   }
-  
+
   for (i in 1 : N) {
     y_hat[i] = dot_product(row(B, county[i]), row(X, i));
   }
-  
+
   y ~ normal(y_hat, sigma);
 }
diff --git a/ARM/Ch.17/radon_multi_varying_coef_17.2.stan b/ARM/Ch.17/radon_multi_varying_coef_17.2.stan
@@ -58,7 +58,7 @@ model {
   }
   
   for (k in 1 : K) {
-    sigma_b[k] = fabs(xi[k]) * sqrt(Sigma_b_raw[k, k]);
+    sigma_b[k] = abs(xi[k]) * sqrt(Sigma_b_raw[k, k]);
     for (k_prime in 1 : K) {
       rho_b[k, k_prime] = Sigma_b_raw[k, k_prime]
                           / sqrt(Sigma_b_raw[k, k]
diff --git a/ARM/Ch.19/pilots_expansion.stan b/ARM/Ch.19/pilots_expansion.stan
@@ -28,7 +28,7 @@ transformed parameters {
   g = xi_g * (g_raw - mean(g_raw));
   d = xi_d * (d_raw - mean(d_raw));
   sigma_g = xi_g * sigma_g_raw;
-  sigma_d = fabs(xi_d) * sigma_d_raw;
+  sigma_d = abs(xi_d) * sigma_d_raw;
   for (i in 1 : N) {
     y_hat[i] = mu + g[treatment[i]] + d[airport[i]];
   }
diff --git a/ARM/Ch.19/schools.stan b/ARM/Ch.19/schools.stan
@@ -14,7 +14,7 @@ transformed parameters {
   vector[N] theta;
   
   theta = mu_theta + xi * eta;
-  sigma_theta = fabs(xi) / sigma_eta;
+  sigma_theta = abs(xi) / sigma_eta;
 }
 model {
   mu_theta ~ normal(0, 100);
diff --git a/BPA/Ch.03/GLM_Poisson.stan b/BPA/Ch.03/GLM_Poisson.stan
@@ -6,7 +6,7 @@ data {
 transformed data {
   vector[n] year_squared;
   vector[n] year_cubed;
-
+  
   year_squared = year .* year;
   year_cubed = year .* year .* year;
 }
@@ -18,18 +18,18 @@ parameters {
 }
 transformed parameters {
   vector[n] log_lambda;
-
+  
   log_lambda = alpha + beta1 * year + beta2 * year_squared
                + beta3 * year_cubed;
 }
 model {
   // Implicit uniform priors are used.
-
+  
   // Likelihood
   C ~ poisson_log(log_lambda);
 }
 generated quantities {
   vector[n] lambda;
-
+  
   lambda = exp(log_lambda);
 }
diff --git a/basic_distributions/triangle.stan b/basic_distributions/triangle.stan
@@ -2,5 +2,5 @@ parameters {
   real<lower=-1, upper=1> y;
 }
 model {
-  target += log1m(fabs(y));
+  target += log1m(abs(y));
 }
diff --git a/bugs_examples/vol1/stacks/stacks_a_normal.stan b/bugs_examples/vol1/stacks/stacks_a_normal.stan
@@ -50,7 +50,7 @@ generated quantities {
   }
   b0 = beta0 - mean_x * b;
   
-  outlier_3 = step(fabs((Y[3] - mu[3]) / sigma) - 2.5);
-  outlier_4 = step(fabs((Y[4] - mu[4]) / sigma) - 2.5);
-  outlier_21 = step(fabs((Y[21] - mu[21]) / sigma) - 2.5);
+  outlier_3 = step(abs((Y[3] - mu[3]) / sigma) - 2.5);
+  outlier_4 = step(abs((Y[4] - mu[4]) / sigma) - 2.5);
+  outlier_21 = step(abs((Y[21] - mu[21]) / sigma) - 2.5);
 }
diff --git a/bugs_examples/vol1/stacks/stacks_b_dexp.stan b/bugs_examples/vol1/stacks/stacks_b_dexp.stan
@@ -52,8 +52,8 @@ generated quantities {
   }
   b0 = beta0 - mean_x * b;
   
-  outlier_1 = step(fabs((Y[1] - mu[1]) / sigma) - 2.5);
-  outlier_3 = step(fabs((Y[3] - mu[3]) / sigma) - 2.5);
-  outlier_4 = step(fabs((Y[4] - mu[4]) / sigma) - 2.5);
-  outlier_21 = step(fabs((Y[21] - mu[21]) / sigma) - 2.5);
+  outlier_1 = step(abs((Y[1] - mu[1]) / sigma) - 2.5);
+  outlier_3 = step(abs((Y[3] - mu[3]) / sigma) - 2.5);
+  outlier_4 = step(abs((Y[4] - mu[4]) / sigma) - 2.5);
+  outlier_21 = step(abs((Y[21] - mu[21]) / sigma) - 2.5);
 }
diff --git a/bugs_examples/vol1/stacks/stacks_c_t4.stan b/bugs_examples/vol1/stacks/stacks_c_t4.stan
@@ -54,7 +54,7 @@ generated quantities {
   }
   b0 = beta0 - mean_x * b;
   
-  outlier_3 = step(fabs((Y[3] - mu[3]) / sigma) - 2.5);
-  outlier_4 = step(fabs((Y[4] - mu[4]) / sigma) - 2.5);
-  outlier_21 = step(fabs((Y[21] - mu[21]) / sigma) - 2.5);
+  outlier_3 = step(abs((Y[3] - mu[3]) / sigma) - 2.5);
+  outlier_4 = step(abs((Y[4] - mu[4]) / sigma) - 2.5);
+  outlier_21 = step(abs((Y[21] - mu[21]) / sigma) - 2.5);
 }
diff --git a/bugs_examples/vol1/stacks/stacks_d_normal_ridge.stan b/bugs_examples/vol1/stacks/stacks_d_normal_ridge.stan
@@ -53,7 +53,7 @@ generated quantities {
   }
   b0 = beta0 - mean_x * b;
   
-  outlier_3 = step(fabs((Y[3] - mu[3]) / sigma) - 2.5);
-  outlier_4 = step(fabs((Y[4] - mu[4]) / sigma) - 2.5);
-  outlier_21 = step(fabs((Y[21] - mu[21]) / sigma) - 2.5);
+  outlier_3 = step(abs((Y[3] - mu[3]) / sigma) - 2.5);
+  outlier_4 = step(abs((Y[4] - mu[4]) / sigma) - 2.5);
+  outlier_21 = step(abs((Y[21] - mu[21]) / sigma) - 2.5);
 }
diff --git a/bugs_examples/vol1/stacks/stacks_e_dexp_ridge.stan b/bugs_examples/vol1/stacks/stacks_e_dexp_ridge.stan
@@ -58,8 +58,8 @@ generated quantities {
   }
   b0 = beta0 - b[1] * mean_x[1] - b[2] * mean_x[2] - b[3] * mean_x[3];
   
-  outlier_1 = step(fabs((Y[1] - mu[1]) / sigma) - 2.5);
-  outlier_3 = step(fabs((Y[3] - mu[3]) / sigma) - 2.5);
-  outlier_4 = step(fabs((Y[4] - mu[4]) / sigma) - 2.5);
-  outlier_21 = step(fabs((Y[21] - mu[21]) / sigma) - 2.5);
+  outlier_1 = step(abs((Y[1] - mu[1]) / sigma) - 2.5);
+  outlier_3 = step(abs((Y[3] - mu[3]) / sigma) - 2.5);
+  outlier_4 = step(abs((Y[4] - mu[4]) / sigma) - 2.5);
+  outlier_21 = step(abs((Y[21] - mu[21]) / sigma) - 2.5);
 }
diff --git a/bugs_examples/vol1/stacks/stacks_f_t4_ridge.stan b/bugs_examples/vol1/stacks/stacks_f_t4_ridge.stan
@@ -57,8 +57,8 @@ generated quantities {
   }
   b0 = beta0 - mean_x * b;
   
-  outlier_1 = step(fabs((Y[3] - mu[3]) / sigma) - 2.5);
-  outlier_3 = step(fabs((Y[3] - mu[3]) / sigma) - 2.5);
-  outlier_4 = step(fabs((Y[4] - mu[4]) / sigma) - 2.5);
-  outlier_21 = step(fabs((Y[21] - mu[21]) / sigma) - 2.5);
+  outlier_1 = step(abs((Y[3] - mu[3]) / sigma) - 2.5);
+  outlier_3 = step(abs((Y[3] - mu[3]) / sigma) - 2.5);
+  outlier_4 = step(abs((Y[4] - mu[4]) / sigma) - 2.5);
+  outlier_21 = step(abs((Y[21] - mu[21]) / sigma) - 2.5);
 }
diff --git a/bugs_examples/vol3/funshapes/squaremc.stan b/bugs_examples/vol3/funshapes/squaremc.stan
@@ -10,8 +10,8 @@
  */
 parameters {
   real<lower=-1, upper=1> x_raw;
-  real<lower=-(1 - sqrt(1 - square(1 - fabs(x_raw)))),
-       upper=(1 - sqrt(1 - square(1 - fabs(x_raw))))> y_raw;
+  real<lower=-(1 - sqrt(1 - square(1 - abs(x_raw)))),
+       upper=(1 - sqrt(1 - square(1 - abs(x_raw))))> y_raw;
 }
 transformed parameters {
   real<lower=-1, upper=1> x;
@@ -20,5 +20,5 @@ transformed parameters {
   y = ((y_raw > 0) ? 1 : -1) - y_raw;
 }
 model {
-  target += log1m(sqrt(1 - square(1 - fabs(x_raw))));
+  target += log1m(sqrt(1 - square(1 - abs(x_raw))));
 }
diff --git a/jupyter/radon/stan/radon_np.stan b/jupyter/radon/stan/radon_np.stan
@@ -1,17 +1,17 @@
 data {
-  int<lower=1> N;  // observations
-  int<lower=1> J;  // counties
+  int<lower=1> N; // observations
+  int<lower=1> J; // counties
   array[N] int<lower=1, upper=J> county;
-  vector[N] x;     // floor
-  vector[N] y;     // radon
+  vector[N] x; // floor
+  vector[N] y; // radon
 }
 parameters {
   vector[J] alpha;
   real beta;
   real<lower=0> sigma;
 }
 model {
-  y ~ normal(alpha[county] + beta * x, sigma);  
+  y ~ normal(alpha[county] + beta * x, sigma);
   alpha ~ normal(0, 10);
   beta ~ normal(0, 10);
   sigma ~ normal(0, 10);
diff --git a/jupyter/radon/stan/radon_np_uranium.stan b/jupyter/radon/stan/radon_np_uranium.stan
@@ -1,10 +1,10 @@
 data {
-  int<lower=1> N;  // observations
-  int<lower=1> J;  // counties
+  int<lower=1> N; // observations
+  int<lower=1> J; // counties
   array[N] int<lower=1, upper=J> county;
-  vector[N] y;     // radon
-  vector[N] x;     // floor
-  vector[N] u;     // uranium
+  vector[N] y; // radon
+  vector[N] x; // floor
+  vector[N] u; // uranium
 }
 parameters {
   vector[J] alpha;
@@ -18,5 +18,6 @@ model {
   sigma ~ normal(0, 10);
 }
 generated quantities {
-  array[N] real y_rep = normal_rng(alpha[county] + beta[1] * x + beta[2] * u, sigma);
+  array[N] real y_rep = normal_rng(alpha[county] + beta[1] * x + beta[2] * u,
+                                   sigma);
 }
diff --git a/jupyter/radon/stan/radon_np_uranium_glm_id_normal.stan b/jupyter/radon/stan/radon_np_uranium_glm_id_normal.stan
@@ -1,10 +1,10 @@
 data {
-  int<lower=1> N;  // observations
-  int<lower=1> J;  // counties
+  int<lower=1> N; // observations
+  int<lower=1> J; // counties
   array[N] int<lower=1, upper=J> county;
-  vector[N] y;     // radon
-  vector[N] x;     // floor
-  vector[N] u;     // uranium
+  vector[N] y; // radon
+  vector[N] x; // floor
+  vector[N] u; // uranium
 }
 transformed data {
   matrix[N, 2] xs = [x', u']';
@@ -18,7 +18,7 @@ transformed parameters {
   vector[N] alphas = alpha[county];
 }
 model {
-  y ~ normal_id_glm(xs, alphas, beta, sigma);  
+  y ~ normal_id_glm(xs, alphas, beta, sigma);
   alpha ~ normal(0, 10);
   beta ~ normal(0, 10);
   sigma ~ normal(0, 10);
diff --git a/jupyter/radon/stan/radon_pp_alpha.stan b/jupyter/radon/stan/radon_pp_alpha.stan
@@ -1,19 +1,19 @@
 data {
-  int<lower=1> N;  // observations
-  int<lower=1> J;  // counties
+  int<lower=1> N; // observations
+  int<lower=1> J; // counties
   array[N] int<lower=1, upper=J> county;
   vector[N] x;
   vector[N] y;
 }
 parameters {
   real mu_alpha;
   real<lower=0> sigma_alpha;
-  vector<offset=mu_alpha, multiplier=sigma_alpha>[J] alpha;  // non-centered parameterization
+  vector<offset=mu_alpha, multiplier=sigma_alpha>[J] alpha; // non-centered parameterization
   real beta;
   real<lower=0> sigma;
 }
 model {
-  y ~ normal(alpha[county] + beta * x, sigma);  
+  y ~ normal(alpha[county] + beta * x, sigma);
   alpha ~ normal(mu_alpha, sigma_alpha); // partial-pooling
   beta ~ normal(0, 10);
   sigma ~ normal(0, 10);
diff --git a/knitr/car-iar-poisson/update_2021_02/bym2.stan b/knitr/car-iar-poisson/update_2021_02/bym2.stan
@@ -69,11 +69,10 @@ parameters {
 }
 transformed parameters {
   // spatial effects (combine heterogeneous and spatially smoothed)
-  vector[I] gamma = sigma * (sqrt(1 - rho) * theta
-			     + sqrt(rho / tau) * phi);
+  vector[I] gamma = sigma * (sqrt(1 - rho) * theta + sqrt(rho / tau) * phi);
 }
 model {
-  y ~ poisson_log(log_E + alpha + x * beta + gamma);  // likelihood
+  y ~ poisson_log(log_E + alpha + x * beta + gamma); // likelihood
   alpha ~ normal(0, 1);
   beta ~ normal(0, 1);
   
diff --git a/knitr/car-iar-poisson/update_2021_02/bym2_islands.stan b/knitr/car-iar-poisson/update_2021_02/bym2_islands.stan
@@ -86,10 +86,11 @@ data {
 transformed data {
   vector[I] log_E = log(E);
   vector[I] taus;
-  for (k in 1:K) {
-    int node_idxs[K_node_cts[k]] = K_node_idxs[k, 1:K_node_cts[k]];
-    for (j in 1:size(node_idxs))
+  for (k in 1 : K) {
+    array[K_node_cts[k]] int node_idxs = K_node_idxs[k, 1 : K_node_cts[k]];
+    for (j in 1 : size(node_idxs)) {
       taus[node_idxs[j]] = tau[k];
+    }
   }
 }
 parameters {
diff --git a/knitr/chapter2/abs_test.stan b/knitr/chapter2/abs_test.stan
@@ -3,7 +3,7 @@ data {
   vector[N] y;
 }
 transformed data {
-  vector[N] abs_y = fabs(y);
+  vector[N] abs_y = abs(y);
 }
 parameters {
   real theta;
diff --git a/knitr/chapter2/worldcup_first_try.stan b/knitr/chapter2/worldcup_first_try.stan
@@ -15,7 +15,7 @@ transformed data {
   vector[N_games] sqrt_dif;
   dif = score_1 - score_2;
   for (i in 1 : N_games) {
-    sqrt_dif[i] = (step(dif[i]) - 0.5) * sqrt(fabs(dif[i]));
+    sqrt_dif[i] = (step(dif[i]) - 0.5) * sqrt(abs(dif[i]));
   }
 }
 parameters {
diff --git a/knitr/chapter2/worldcup_first_try_noprior.stan b/knitr/chapter2/worldcup_first_try_noprior.stan
@@ -16,7 +16,7 @@ transformed data {
   vector[N_games] sqrt_dif;
   dif = score_1 - score_2;
   for (i in 1 : N_games) {
-    sqrt_dif[i] = (step(dif[i]) - 0.5) * sqrt(fabs(dif[i]));
+    sqrt_dif[i] = (step(dif[i]) - 0.5) * sqrt(abs(dif[i]));
   }
 }
 parameters {
diff --git a/knitr/chapter2/worldcup_fixed.stan b/knitr/chapter2/worldcup_fixed.stan
@@ -15,7 +15,7 @@ transformed data {
   vector[N_games] sqrt_dif;
   dif = score_1 - score_2;
   for (i in 1 : N_games) {
-    sqrt_dif[i] = 2 * (step(dif[i]) - 0.5) * sqrt(fabs(dif[i]));
+    sqrt_dif[i] = 2 * (step(dif[i]) - 0.5) * sqrt(abs(dif[i]));
   }
 }
 parameters {
@@ -38,5 +38,5 @@ generated quantities {
   for (n in 1 : N_games) {
     y_rep[n] = student_t_rng(df, a[team_1[n]] - a[team_2[n]], sigma_y);
   }
-  y_rep_original_scale = y_rep .* fabs(y_rep);
+  y_rep_original_scale = y_rep .* abs(y_rep);
 }
diff --git a/knitr/chapter2/worldcup_with_replication.stan b/knitr/chapter2/worldcup_with_replication.stan
@@ -15,7 +15,7 @@ transformed data {
   vector[N_games] sqrt_dif;
   dif = score_1 - score_2;
   for (i in 1 : N_games) {
-    sqrt_dif[i] = (step(dif[i]) - 0.5) * sqrt(fabs(dif[i]));
+    sqrt_dif[i] = (step(dif[i]) - 0.5) * sqrt(abs(dif[i]));
   }
 }
 parameters {
@@ -38,5 +38,5 @@ generated quantities {
   for (n in 1 : N_games) {
     y_rep[n] = student_t_rng(df, a[team_1[n]] - a[team_2[n]], sigma_y);
   }
-  y_rep_original_scale = y_rep .* fabs(y_rep);
+  y_rep_original_scale = y_rep .* abs(y_rep);
 }
diff --git a/knitr/pest-control/stan_programs/multiple_NB_regression_dgp.stan b/knitr/pest-control/stan_programs/multiple_NB_regression_dgp.stan
@@ -12,7 +12,7 @@ generated quantities {
   real alpha = normal_rng(log(4), 0.1);
   real beta = normal_rng(-0.25, 0.1);
   real beta_super = normal_rng(-0.5, 0.1);
-  real inv_phi = fabs(normal_rng(0, 1));
+  real inv_phi = abs(normal_rng(0, 1));
   
   for (n in 1 : N) {
     log_sq_foot[n] = normal_rng(1.5, 0.1);
diff --git a/knitr/world-cup/abs_test.stan b/knitr/world-cup/abs_test.stan
diff --git a/knitr/world-cup/worldcup_first_try.stan b/knitr/world-cup/worldcup_first_try.stan
diff --git a/knitr/world-cup/worldcup_first_try_noprior.stan b/knitr/world-cup/worldcup_first_try_noprior.stan
diff --git a/knitr/world-cup/worldcup_fixed.stan b/knitr/world-cup/worldcup_fixed.stan
diff --git a/knitr/world-cup/worldcup_with_replication.stan b/knitr/world-cup/worldcup_with_replication.stan
diff --git a/regression_tests/mother.stan b/regression_tests/mother.stan

Original file line number	Diff line number	Diff line change
`@@ -58,7 +58,7 @@ model {`
`58`	`58`	`}`
`59`	`59`
`60`	`60`	`for (k in 1 : K) {`
`61`		`- sigma_b[k] = fabs(xi[k]) * sqrt(Sigma_b_raw[k, k]);`
	`61`	`+ sigma_b[k] = abs(xi[k]) * sqrt(Sigma_b_raw[k, k]);`
`62`	`62`	`for (k_prime in 1 : K) {`
`63`	`63`	`rho_b[k, k_prime] = Sigma_b_raw[k, k_prime]`
`64`	`64`	`/ sqrt(Sigma_b_raw[k, k]`
Original file line number	Diff line number	Diff line change
`@@ -28,7 +28,7 @@ transformed parameters {`
`28`	`28`	`g = xi_g * (g_raw - mean(g_raw));`
`29`	`29`	`d = xi_d * (d_raw - mean(d_raw));`
`30`	`30`	`sigma_g = xi_g * sigma_g_raw;`
`31`		`- sigma_d = fabs(xi_d) * sigma_d_raw;`
	`31`	`+ sigma_d = abs(xi_d) * sigma_d_raw;`
`32`	`32`	`for (i in 1 : N) {`
`33`	`33`	`y_hat[i] = mu + g[treatment[i]] + d[airport[i]];`
`34`	`34`	`}`
Original file line number	Diff line number	Diff line change
`@@ -14,7 +14,7 @@ transformed parameters {`
`14`	`14`	`vector[N] theta;`
`15`	`15`
`16`	`16`	`theta = mu_theta + xi * eta;`
`17`		`- sigma_theta = fabs(xi) / sigma_eta;`
	`17`	`+ sigma_theta = abs(xi) / sigma_eta;`
`18`	`18`	`}`
`19`	`19`	`model {`
`20`	`20`	`mu_theta ~ normal(0, 100);`
Original file line number	Diff line number	Diff line change
`@@ -2,5 +2,5 @@ parameters {`
`2`	`2`	`real<lower=-1, upper=1> y;`
`3`	`3`	`}`
`4`	`4`	`model {`
`5`		`- target += log1m(fabs(y));`
	`5`	`+ target += log1m(abs(y));`
`6`	`6`	`}`
Original file line number	Diff line number	Diff line change
`@@ -50,7 +50,7 @@ generated quantities {`
`50`	`50`	`}`
`51`	`51`	`b0 = beta0 - mean_x * b;`
`52`	`52`
`53`		`- outlier_3 = step(fabs((Y[3] - mu[3]) / sigma) - 2.5);`
`54`		`- outlier_4 = step(fabs((Y[4] - mu[4]) / sigma) - 2.5);`
`55`		`- outlier_21 = step(fabs((Y[21] - mu[21]) / sigma) - 2.5);`
	`53`	`+ outlier_3 = step(abs((Y[3] - mu[3]) / sigma) - 2.5);`
	`54`	`+ outlier_4 = step(abs((Y[4] - mu[4]) / sigma) - 2.5);`
	`55`	`+ outlier_21 = step(abs((Y[21] - mu[21]) / sigma) - 2.5);`
`56`	`56`	`}`
Original file line number	Diff line number	Diff line change
`@@ -52,8 +52,8 @@ generated quantities {`
`52`	`52`	`}`
`53`	`53`	`b0 = beta0 - mean_x * b;`
`54`	`54`
`55`		`- outlier_1 = step(fabs((Y[1] - mu[1]) / sigma) - 2.5);`
`56`		`- outlier_3 = step(fabs((Y[3] - mu[3]) / sigma) - 2.5);`
`57`		`- outlier_4 = step(fabs((Y[4] - mu[4]) / sigma) - 2.5);`
`58`		`- outlier_21 = step(fabs((Y[21] - mu[21]) / sigma) - 2.5);`
	`55`	`+ outlier_1 = step(abs((Y[1] - mu[1]) / sigma) - 2.5);`
	`56`	`+ outlier_3 = step(abs((Y[3] - mu[3]) / sigma) - 2.5);`
	`57`	`+ outlier_4 = step(abs((Y[4] - mu[4]) / sigma) - 2.5);`
	`58`	`+ outlier_21 = step(abs((Y[21] - mu[21]) / sigma) - 2.5);`
`59`	`59`	`}`
Original file line number	Diff line number	Diff line change
`@@ -54,7 +54,7 @@ generated quantities {`
`54`	`54`	`}`
`55`	`55`	`b0 = beta0 - mean_x * b;`
`56`	`56`
`57`		`- outlier_3 = step(fabs((Y[3] - mu[3]) / sigma) - 2.5);`
`58`		`- outlier_4 = step(fabs((Y[4] - mu[4]) / sigma) - 2.5);`
`59`		`- outlier_21 = step(fabs((Y[21] - mu[21]) / sigma) - 2.5);`
	`57`	`+ outlier_3 = step(abs((Y[3] - mu[3]) / sigma) - 2.5);`
	`58`	`+ outlier_4 = step(abs((Y[4] - mu[4]) / sigma) - 2.5);`
	`59`	`+ outlier_21 = step(abs((Y[21] - mu[21]) / sigma) - 2.5);`
`60`	`60`	`}`
Original file line number	Diff line number	Diff line change
`@@ -53,7 +53,7 @@ generated quantities {`
`53`	`53`	`}`
`54`	`54`	`b0 = beta0 - mean_x * b;`
`55`	`55`
`56`		`- outlier_3 = step(fabs((Y[3] - mu[3]) / sigma) - 2.5);`
`57`		`- outlier_4 = step(fabs((Y[4] - mu[4]) / sigma) - 2.5);`
`58`		`- outlier_21 = step(fabs((Y[21] - mu[21]) / sigma) - 2.5);`
	`56`	`+ outlier_3 = step(abs((Y[3] - mu[3]) / sigma) - 2.5);`
	`57`	`+ outlier_4 = step(abs((Y[4] - mu[4]) / sigma) - 2.5);`
	`58`	`+ outlier_21 = step(abs((Y[21] - mu[21]) / sigma) - 2.5);`
`59`	`59`	`}`