Treść książki
Przejdź do opcji czytnikaPrzejdź do nawigacjiPrzejdź do informacjiPrzejdź do stopki
AlicjaGanczarek-Gamrot
ASSESSINGLONGMEMORY
CHARACTERISTICS
OFENERGYPRICES
Introduction
Timeseriesfromfinancialandcommoditymarketsarecharacterizedby
verystrongautocorrelation.Significantvalueofautocorrelationcoefficientfor
largelagindicatesthelongmemoryintimeseries.Oftentimeseries,long
memoryisthecauseofnonstationarity.Thenoftentimeserieshaveunitroots.
Theunitrootcanbeeliminatedfromtimeseriesthroughintegerdifferencing.
But
alotofempiricalresultsshowthatintegerdifferencingofempiricaltimeseries
isunnecessary.Veryoftendegreeofdifferenceoftimeseriesisrealandsmaller
thanone.Thentimeseriesarestationary.
Inthispapertheresultsofestimatinglongmemoryeffectsintimeserieson
Polishelectricenergymarketarepresented.Methods,whicharedescribedby
ZivotandWang
1areusedtoestimatelongmemoryeffects.
1.Longmemory
Thestationarytimeseriesythaslongmemoryorlongrangedependenceif:
ρ
(
k
)
k
→
→
∞
C
ρ
k
−
α
,
(1)
1E.Zivot,J.Wang:ModelingFinancialTimeSerieswithS-PLUS.Springer,NewYork2006.
