Merge 5a95f33 into 25e9ef3

.travis.yml

@@ -4,7 +4,8 @@ os:
   - linux
   - osx
 julia:
-  - 0.6
+  - 0.7
+  - 1.0
   - nightly
 notifications:
   email: false

REQUIRE

@@ -1,2 +1,3 @@
-julia 0.6
+julia 0.7
 Distributions
+SpecialFunctions

src/MicrostructureNoise.jl

@@ -1,6 +1,8 @@
 module MicrostructureNoise
 
-using Distributions
+using SpecialFunctions, Distributions
+
+using Statistics
 
 include("microstructure.jl")
 

src/microstructure.jl

@@ -120,7 +120,7 @@ function MCMC(Π::Union{Prior,Dict}, tt, y, α0::Float64, σα, iterations; subi
 
 
     Z = zeros(N)
-    ii = Vector(N) # vector of start indices of increments
+    ii = Vector(undef, N) # vector of start indices of increments
     td = zeros(N+1)
     for k in 1:N
         if k == N
@@ -139,9 +139,7 @@ function MCMC(Π::Union{Prior,Dict}, tt, y, α0::Float64, σα, iterations; subi
     acc = 0
     αs = Float64[]
     si = 1
-    fixalpha = false
-    fixeta = false
-
+
     μ = zeros(n+1)
     C = zeros(n+1)
     θs = Any[]
@@ -151,7 +149,7 @@ function MCMC(Π::Union{Prior,Dict}, tt, y, α0::Float64, σα, iterations; subi
 
     for iter in 1:iterations
         # update Zk (necessary because x changes)
-        if !(eta == 0.0 && fixeta)
+        if !(η == 0.0 && fixeta)
             for k in 1:N
                 Z[k] = sum((x[i+1] - x[i]).^2 ./ (tt[i+1]-tt[i]) for i in ii[k])
             end
@@ -194,7 +192,7 @@ function MCMC(Π::Union{Prior,Dict}, tt, y, α0::Float64, σα, iterations; subi
         end
         if !fixeta
             # update eta
-            assert(length(x) == n + 1)
+            @assert(length(x) == n + 1)
             z = sum((x[i] - y[i])^2 for i in 2:(n+1))
             η = rand(InverseGamma(αη + n/2, βη + z/2))
 
@@ -207,7 +205,7 @@ function MCMC(Π::Union{Prior,Dict}, tt, y, α0::Float64, σα, iterations; subi
         # Sample x from Kalman smoothing distribution
 
         # Forward pass
-        if eta == 0.0 && fixeta
+        if η == 0.0 && fixeta
             # do nothing
         else 
             C[1] = C0
@@ -254,7 +252,7 @@ function MCMC(Π::Union{Prior,Dict}, tt, y, α0::Float64, σα, iterations; subi
         end
     end
 
-    td, samples, ηs, αs, round(acc/iterations, 3)
+    td, samples, ηs, αs, round(acc/iterations, digits=3)
 end
 
 """
@@ -293,11 +291,11 @@ posterior median and mean of `s^2`, as well as posterior mean of `s`.
 function posterior_volatility(td, samples; burnin = size(samples, 2)÷3, qu = 0.90)
     p = 1.0 - qu 
     A = view(samples, :, burnin:size(samples, 2))
-    post_qup = mapslices(v-> quantile(v, 1 - p/2), A, 2)
-    post_mean = mean(A, 2)
-    post_mean_root = mean(sqrt.(A), 2)
-    post_median = median(A, 2)
-    post_qlow = mapslices(v-> quantile(v,  p/2), A, 2)
+    post_qup = mapslices(v-> quantile(v, 1 - p/2), A, dims=2)
+    post_mean = mean(A, dims=2)
+    post_mean_root = mean(sqrt.(A), dims=2)
+    post_median = median(A, dims=2)
+    post_qlow = mapslices(v-> quantile(v,  p/2), A, dims=2)
     Posterior(
         td,
         post_qlow,

test/piecewise.jl

@@ -1,3 +1,3 @@
-using Base.Test, MicrostructureNoise
+using Test, MicrostructureNoise
 
 @test MicrostructureNoise.piecewise([0,1,2],[10,11]) == ([0, 1, 1, 2], [10, 10, 11, 11])

test/runtests.jl

@@ -1,5 +1,5 @@
 using MicrostructureNoise
-using Base.Test
+using Test
 
 include("smoketest.jl")
 include("piecewise.jl")

test/smoketest.jl

@@ -1,5 +1,8 @@
-using MicrostructureNoise, Distributions, Base.Test
-srand(10)
+using MicrostructureNoise, Distributions
+
+using Random, LinearAlgebra
+
+Random.seed!(10)
 η = 0.01
 
 n = 1000
@@ -35,4 +38,4 @@ posterior = MicrostructureNoise.posterior_volatility(td, θs)
 # Test posterior mean versus truth
 
 @test abs(mean(ηs)-η) < 0.2*η
-@test norm(diff(td).*(mean(θs,2) - σ.(td[1:end-1]).^2)) < 1
+@test norm(diff(td).*(mean(θs, dims=2) - σ.(td[1:end-1]).^2)) < 1