Symmetrise and whitespace

Author: Andrew Johnson

basic_distributions/binormal.stan

@@ -7,5 +7,5 @@ parameters {
     vector[2] y;
 }
 model {
-      y ~ multi_normal(mu,Sigma);
+      y ~ multi_normal(mu, Sigma);
 }

basic_distributions/normal.stan

@@ -2,5 +2,5 @@ parameters {
   real y;
 }
 model {
-  y ~ normal(0,1);
+  y ~ normal(0, 1);
 }

basic_distributions/ordered_pair.stan

@@ -3,10 +3,10 @@ transformed data {
     real H = 5;
 }
 parameters {
-    real<lower=L,upper=H> a;
-    real<lower=a,upper=H> b;
+    real<lower=L, upper=H> a;
+    real<lower=a, upper=H> b;
 }
 model {
-//    a ~ uniform(L,b);
-//    b ~ uniform(a,H);
+//    a ~ uniform(L, b);
+//    b ~ uniform(a, H);
 }

basic_distributions/sorted_pair.stan

@@ -1,10 +1,10 @@
 parameters {
-    real<lower=-1,upper=1> x1;
-    real<lower=-1,upper=1> x2;
+    real<lower=-1, upper=1> x1;
+    real<lower=-1, upper=1> x2;
 }
 model {
 }
 generated quantities {
-    real a = fmax(x1,x2);
-    real b = fmin(x1,x2);
+    real a = fmax(x1, x2);
+    real b = fmin(x1, x2);
 }

basic_distributions/triangle.stan

@@ -1,5 +1,5 @@
 parameters {
-  real<lower=-1,upper=1> y;
+  real<lower=-1, upper=1> y;
 }
 model {
   target += log1m(fabs(y));

basic_distributions/uniform.stan

@@ -1,6 +1,6 @@
 parameters {
-    real<lower=0,upper=1> y;
+    real<lower=0, upper=1> y;
 }
 model {
-    y ~ uniform(0,1);
+    y ~ uniform(0, 1);
 }

basic_distributions/wishart2.stan

@@ -2,19 +2,17 @@ transformed data {
   cov_matrix[4] S;
 
   S[1,1] = 2.9983662;
-  S[1,2] = 0.2898776;
-  S[1,3] = -2.650523;
-  S[1,4] = 0.1055911;
+  S[2,1] = 0.2898776;
+  S[3,1] = -2.650523;
+  S[4,1] = 0.1055911;
   S[2,2] = 11.4803610;
-  S[2,3] = 7.157993;
-  S[2,4] = -3.1129955;
+  S[3,2] = 7.157993;
+  S[4,2] = -3.1129955;
   S[3,3] = 11.676181;
-  S[3,4] = -3.5866852;
+  S[4,3] = -3.5866852;
   S[4,4] = 1.4482736;
 
-  for (m in 1:4)
-    for (n in 1:(m-1))
-      S[m,n] = S[n,m];
+  S = symmetrize_from_lower_tri(S);
 }
 parameters {
   cov_matrix[4] W; 

basic_distributions/wishart2x2.stan

@@ -21,8 +21,8 @@ parameters {
 transformed parameters {
   real rho = tanh(x);
   real cov = rho * sd1 * sd2;
-  matrix[2,2] W = [[sd1^2, cov],
-                   [cov, sd2^2]];
+  matrix[2,2] W = [[sd1 ^ 2, cov],
+                   [cov, sd2 ^ 2]];
 }
 model {
   // apply log Jacobian determinant of transform: