introduce view in LTS and LMS

Author: jbytecode

CHANGELOG.md

@@ -1,6 +1,7 @@
 # v0.11.5 (Upcoming Release)
 
 - Initial implementation of the robust hat matrix regression estimator
+- Introduce `view` in LTS and LMS.
 
 # v0.11.4 
 

src/lms.jl

@@ -79,7 +79,7 @@ function lms(X::AbstractMatrix{Float64}, y::AbstractVector{Float64}; iters = not
         try
             k = rand(kindices)
             sampledindices = sample(indices, k, replace = false)
-            betas = X[sampledindices, :] \ y[sampledindices]
+            betas = view(X, sampledindices, :) \ view(y, sampledindices)
             res = sort!((y .- X * betas) .^ 2.0)
             m2 = res[h]
             if m2 < bestobjective

src/lts.jl

@@ -51,11 +51,11 @@ function iterateCSteps(
     iter::Int = 0
     sortedresindices = Array{Int}(undef, n)
     while iter < maxiter
-        tempols = ols(X[subsetindices, :], y[subsetindices])
+        tempols = ols(view(X, subsetindices, :), view(y, subsetindices))
         res = y - X * coef(tempols)
         sortperm!(sortedresindices, abs.(res))
-        subsetindices = sortedresindices[1:h]
-        objective = sum(sort!(res .^ 2.0)[1:h])
+        subsetindices = view(sortedresindices, 1:h)
+        objective = sum(view(sort!(res .^ 2.0), 1:h))
         if isapprox(oldobjective, objective, atol=eps)
             break
         end
@@ -172,7 +172,7 @@ function lts(X::AbstractMatrix{Float64}, y::AbstractVector{Float64}; iters=nothi
         end
     end
 
-    ltsbetas = X[besthsubset, :] \ y[besthsubset]
+    ltsbetas = view(X, besthsubset, :) \ view(y, besthsubset)
     ltsres = y - X * ltsbetas
     ltsS = sqrt(sum((ltsres .^ 2.0)[1:h]) / (h - p))
     ltsresmean = mean(ltsres[besthsubset])

test/runtests.jl

@@ -6,9 +6,9 @@ using LinearAlgebra
 using LinRegOutliers
 import Plots: RGBX
 
-include("testdiagnostics.jl")
 include("testbasis.jl")
 include("testols.jl")
+include("testdiagnostics.jl")
 include("tesths93.jl")
 include("testks89.jl")
 include("testsmr98.jl")