Compare ⋅ s-broda/ARCHModels.jl

#dimensional array of the right type.
@inline function loglik!(ht::AbstractVector{T2}, lht::AbstractVector{T2},
                         zt::AbstractVector{T2}, at::AbstractVector{T2}, vs::Type{VS}, ::Type{SD}, meanspec::MS,
                         data::Vector{T1}, coefs::AbstractVector{T3}, subsetmask=trues(nparams(vs))
                         data::Vector{T1}, coefs::AbstractVector{T3}, subsetmask=trues(nparams(vs)), returnearly=false
                         ) where {VS<:UnivariateVolatilitySpec, SD<:StandardizedDistribution,
                                  MS<:MeanSpec, T1<:AbstractFloat, T2, T3
                                  }
    garchcoefs, distcoefs, meancoefs = splitcoefs(coefs, VS, SD, meanspec)
	lowergarch, uppergarch = constraints(VS, T1)
	lowerdist, upperdist = constraints(SD, T1)
    lowermean, uppermean = constraints(MS, T1)
    all(lowerdist.<distcoefs.<upperdist) && all(lowermean.<meancoefs.<uppermean) && all(lowergarch[subsetmask].<garchcoefs[subsetmask].<uppergarch[subsetmask]) || return T2(-Inf)
	all_inbounds = all(lowerdist.<distcoefs.<upperdist) && all(lowermean.<meancoefs.<uppermean) && all(lowergarch[subsetmask].<garchcoefs[subsetmask].<uppergarch[subsetmask])
    returnearly && !all_inbounds && return T2(-Inf)
	garchcoefs .*= subsetmask
    T = length(data)
	r1 = presample(VS)
	r2 = presample(meanspec)
    r = max(r1, r2)
    T > r || error("Sample too small.")
    T - r > 0 || error("Sample too small.")
	ki = kernelinvariants(SD, distcoefs)
    @inbounds begin
        h0 = var(data) # could be moved outside

@@ -329,10 +330,11 @@

Loading

        end#for
    end#inbounds
    LL += T*logconst(SD, distcoefs)
	return all_inbounds ? LL : T2(-Inf)
end#function

function loglik(spec::Type{VS}, dist::Type{SD}, meanspec::MS,
                   data::Vector{<:AbstractFloat}, coefs::AbstractVector{T2}, subsetmask=trues(nparams(spec))
                   data::Vector{<:AbstractFloat}, coefs::AbstractVector{T2}, subsetmask=trues(nparams(spec)), returnearly=false
                   ) where {VS<:UnivariateVolatilitySpec, SD<:StandardizedDistribution,
                            MS<:MeanSpec, T2
                            }

@@ -342,7 +344,7 @@

Loading

    lht = CircularBuffer{T2}(r)
    zt = CircularBuffer{T2}(r)
	at = CircularBuffer{T2}(r)
    loglik!(ht, lht, zt, at, spec, dist, meanspec, data, coefs, subsetmask)
    loglik!(ht, lht, zt, at, spec, dist, meanspec, data, coefs, subsetmask, returnearly)

end


@@ -375,7 +377,7 @@

Loading

              ) where {VS<:UnivariateVolatilitySpec, SD<:StandardizedDistribution,
                       MS<:MeanSpec, T<:AbstractFloat
                       }
    obj = x -> -loglik(VS, SD, meanspec, data, x)
    obj = x -> -loglik(VS, SD, meanspec, data, x, trues(length(garchcoefs)), true)
    coefs = vcat(garchcoefs, distcoefs, meancoefs)
    res = optimize(obj, coefs, algorithm; autodiff=autodiff, kwargs...)
    coefs .= Optim.minimizer(res)

@@ -458,7 +460,7 @@

Loading

	distcoefs = startingvals(SD, data)
    meancoefs = startingvals(ms, data)

	obj = x -> -loglik(VS_large, SD, ms, data, x, mask)
	obj = x -> -loglik(VS_large, SD, ms, data, x, mask, true)
    coefs = vcat(garchcoefs, distcoefs, meancoefs)
    res = optimize(obj, coefs, algorithm; autodiff=autodiff, kwargs...)
    coefs .= Optim.minimizer(res)

@@ -511,7 +513,7 @@

Loading

- `algorithm=BFGS(), autodiff=:forward, kwargs...`: passed on to the optimizer.

# Example
```jldoctest
```
julia> selectmodel(EGARCH, BG96)

EGARCH{1, 1, 2} model with Gaussian errors, T=1974.

@@ -171,14 +171,17 @@

Loading

    return lower, upper
end

function startingvals(::ARMA{p, q, T}, data::Vector{T})  where {p, q, T<:AbstractFloat}
function startingvals(mod::ARMA{p, q, T}, data::Vector{T})  where {p, q, T<:AbstractFloat}
    N = length(data)
    X = Matrix{T}(undef, N-p, p+1)
    X[:, 1] .= T(1)
    for i = 1:p
        X[:, i+1] .= data[p-i+1:N-i]
    end
    phi = X \ data[p+1:end]
    lower, upper = constraints(ARMA{p, q}, T)
    phi[2:end] .= max.(phi[2:end], lower[2:p+1]*.99)
    phi[2:end] .= min.(phi[2:end], upper[2:p+1]*.99)
    return T[phi..., zeros(T, q)...]
end


@@ -44,7 +44,7 @@

Loading

    Ji = try
        inv(J)
    catch e
        if e isa LinearAlgebra.SingularException
        if e in [LinearAlgebra.SingularException, LinearAlgebra.LAPACKException(1)]
            @warn "Fisher information is singular; vcov matrix is inaccurate."
            pinv(J)
        else

Everything is accounted for!

No changes detected that need to be reviewed.

What changes does Codecov check for?

Lines, not adjusted in diff, that have changed coverage data.

Files that introduced coverage data that had none before.

Files that have missing coverage data that once were tracked.

e7810d8

+5

99.41% 100.00% +<.01%

add convenience functions for ARMA and bump version

s-broda

99b389f

CI Passed

Files		•	•	•	Coverage
src	+5	+5	ø	ø	^+<.01% 99.41%
Project Totals (12 files)	1,196	1,189	0	7	99.41%

287	287		#dimensional array of the right type.
288	288		@inline function loglik!(ht::AbstractVector{T2}, lht::AbstractVector{T2},
289	289		zt::AbstractVector{T2}, at::AbstractVector{T2}, vs::Type{VS}, ::Type{SD}, meanspec::MS,
290		-	data::Vector{T1}, coefs::AbstractVector{T3}, subsetmask=trues(nparams(vs))
	290	+	data::Vector{T1}, coefs::AbstractVector{T3}, subsetmask=trues(nparams(vs)), returnearly=false
291	291		) where {VS<:UnivariateVolatilitySpec, SD<:StandardizedDistribution,
292	292		MS<:MeanSpec, T1<:AbstractFloat, T2, T3
293	293		}
294	294		garchcoefs, distcoefs, meancoefs = splitcoefs(coefs, VS, SD, meanspec)
295	295		lowergarch, uppergarch = constraints(VS, T1)
296	296		lowerdist, upperdist = constraints(SD, T1)
297	297		lowermean, uppermean = constraints(MS, T1)
298		-	all(lowerdist.<distcoefs.<upperdist) && all(lowermean.<meancoefs.<uppermean) && all(lowergarch[subsetmask].<garchcoefs[subsetmask].<uppergarch[subsetmask]) \|\| return T2(-Inf)
	298	+	all_inbounds = all(lowerdist.<distcoefs.<upperdist) && all(lowermean.<meancoefs.<uppermean) && all(lowergarch[subsetmask].<garchcoefs[subsetmask].<uppergarch[subsetmask])
	299	+	returnearly && !all_inbounds && return T2(-Inf)
299	300		garchcoefs .*= subsetmask
300	301		T = length(data)
301	302		r1 = presample(VS)
302	303		r2 = presample(meanspec)
303	304		r = max(r1, r2)
304		-	T > r \|\| error("Sample too small.")
	305	+	T - r > 0 \|\| error("Sample too small.")
305	306		ki = kernelinvariants(SD, distcoefs)
306	307		@inbounds begin
307	308		h0 = var(data) # could be moved outside

329	330		end#for
330	331		end#inbounds
331	332		LL += T*logconst(SD, distcoefs)
	333	+	return all_inbounds ? LL : T2(-Inf)
332	334		end#function
333	335
334	336		function loglik(spec::Type{VS}, dist::Type{SD}, meanspec::MS,
335		-	data::Vector{<:AbstractFloat}, coefs::AbstractVector{T2}, subsetmask=trues(nparams(spec))
	337	+	data::Vector{<:AbstractFloat}, coefs::AbstractVector{T2}, subsetmask=trues(nparams(spec)), returnearly=false
336	338		) where {VS<:UnivariateVolatilitySpec, SD<:StandardizedDistribution,
337	339		MS<:MeanSpec, T2
338	340		}

342	344		lht = CircularBuffer{T2}(r)
343	345		zt = CircularBuffer{T2}(r)
344	346		at = CircularBuffer{T2}(r)
345		-	loglik!(ht, lht, zt, at, spec, dist, meanspec, data, coefs, subsetmask)
	347	+	loglik!(ht, lht, zt, at, spec, dist, meanspec, data, coefs, subsetmask, returnearly)
346	348
347	349		end
348	350

375	377		) where {VS<:UnivariateVolatilitySpec, SD<:StandardizedDistribution,
376	378		MS<:MeanSpec, T<:AbstractFloat
377	379		}
378		-	obj = x -> -loglik(VS, SD, meanspec, data, x)
	380	+	obj = x -> -loglik(VS, SD, meanspec, data, x, trues(length(garchcoefs)), true)
379	381		coefs = vcat(garchcoefs, distcoefs, meancoefs)
380	382		res = optimize(obj, coefs, algorithm; autodiff=autodiff, kwargs...)
381	383		coefs .= Optim.minimizer(res)

458	460		distcoefs = startingvals(SD, data)
459	461		meancoefs = startingvals(ms, data)
460	462
461		-	obj = x -> -loglik(VS_large, SD, ms, data, x, mask)
	463	+	obj = x -> -loglik(VS_large, SD, ms, data, x, mask, true)
462	464		coefs = vcat(garchcoefs, distcoefs, meancoefs)
463	465		res = optimize(obj, coefs, algorithm; autodiff=autodiff, kwargs...)
464	466		coefs .= Optim.minimizer(res)

Compare `99b389f ... +0 ... e7810d8`

No flags found

Everything is accounted for!

What changes does Codecov check for?

2 Commits

511	513		- `algorithm=BFGS(), autodiff=:forward, kwargs...`: passed on to the optimizer.
512	514
513	515		# Example
514		-	```jldoctest
	516	+	```
515	517		julia> selectmodel(EGARCH, BG96)
516	518
517	519		EGARCH{1, 1, 2} model with Gaussian errors, T=1974.

171	171		return lower, upper
172	172		end
173	173
174		-	function startingvals(::ARMA{p, q, T}, data::Vector{T}) where {p, q, T<:AbstractFloat}
	174	+	function startingvals(mod::ARMA{p, q, T}, data::Vector{T}) where {p, q, T<:AbstractFloat}
175	175		N = length(data)
176	176		X = Matrix{T}(undef, N-p, p+1)
177	177		X[:, 1] .= T(1)
178	178		for i = 1:p
179	179		X[:, i+1] .= data[p-i+1:N-i]
180	180		end
181	181		phi = X \ data[p+1:end]
	182	+	lower, upper = constraints(ARMA{p, q}, T)
	183	+	phi[2:end] .= max.(phi[2:end], lower[2:p+1]*.99)
	184	+	phi[2:end] .= min.(phi[2:end], upper[2:p+1]*.99)
182	185		return T[phi..., zeros(T, q)...]
183	186		end
184	187

44	44		Ji = try
45	45		inv(J)
46	46		catch e
47		-	if e isa LinearAlgebra.SingularException
	47	+	if e in [LinearAlgebra.SingularException, LinearAlgebra.LAPACKException(1)]
48	48		@warn "Fisher information is singular; vcov matrix is inaccurate."
49	49		pinv(J)
50	50		else

Navigation	Overlay
`t` Navigate files	`h` Toggle hits
`y` Change url to tip of branch	`m` Toggle misses
`b / v` Jump to prev/next hit line	`p` Toggle partial
`z / x` Jump to prev/next missed or partial line	`1..9` Toggle flags
`shift + o` Open current page in GitHub	`a` Toggle all on
`/ or ?` Show keyboard shortcuts dialog	`c` Toggle context lines or commits

Compare 99b389f ... +0 ... e7810d8

No flags found

Everything is accounted for!

What changes does Codecov check for?

2 Commits

Compare `99b389f ... +0 ... e7810d8`