Exponential power model



   [exppower_ex1]
   
   model
   {
      for( i in 1 : N )
      {
      x[i] ~ dexp.power(alpha, lambda)
      }
      
   # Prior distributions of the model parameters   
   
         alpha ~ dunif(0, 20.0)
         lambda~ dunif(0, 1.0)      
   }

Simulated data set with alpha = 2.5 and lambda = 0.25
The MLEs are alpha = 2.5920853 and lambda = 0.2042697

Data
list(N=33, x=c(0.7924519, 1.3798784, 1.7264492, 1.9919010, 2.2133770, 2.4065532, 2.5797627, 2.7380554, 2.8847681, 3.0222479, 3.1522276, 3.2760370, 3.3947323, 3.5091779, 3.6201017, 3.7281341, 3.8338369, 3.9377261, 4.0402905, 4.1420092, 4.2433687, 4.3448821, 4.4471122, 4.5507015, 4.6564147, 4.7652034, 4.8783082, 4.9974344, 5.1250778, 5.2651932, 5.4247750, 5.6185011, 5.8890629))
Inits for chain 1
list(alpha=2.0, lambda=0.1)
   
Inits for chain 2
list(alpha=1.0, lambda=0.2)



Results


[exppower_ex2]

MAP estimates are

[exppower_ex3]

On this page