CS代考 Times h Y la l it ilee YE

Times h Y la l it ilee YE
a sequence f
We observe realization f e time series
4 9 Two assumptions regarding fine series

Copyright By PowCoder代写 加微信 powcoder

stationerity
Ergodicity
STATWNAYIje.ly
random variables
alt called random process of stochastic process
doesn’t Mean variances covariances
A process whose behavior
the joint pdf of all combinations of RVs remain
the same 9 Y
isthesameas Pdf
useful property
f 43yoyo is thesameor
change one skewness kurtosis
any function t Ye y 142 is also
strictly stationery stationery

Yt is on iid process the Lye a strictlystationery independent
identically distributed
five periods fr all t
Covariance correlation between random variables for

is called the autocovariance
is called the autocorrelation
mean is constant over all Elie u
all five periods Ye It for all t
time periods
between them
constant over
depends on the timedifference tI
function for t.ie function for time
difference
difference

IssianWHITE.no
Yt 7119 then444 is called GaussianWhitehorse
All Ye have the same ween
All Ye have the
Eye is a ISIMULATIINRI
same variance 6 are independent
compounded monthly returnon stock
Example return generating process
represents the
continuously
Ld N 1SiMOLATEIN
0.01 0.05 0001
0M.0521 terms
In co theniEcy_qyqpenlywrorktre.c.n
EItE ECEELIDOU2
beam they’re independent

ret retrei
I 6 My relay is a
INDEPENDENT WHITE
anydistribution
E YeN IWNfo
WHITE NOISE
tf Mons a with
t distribution
3 degrees of Medan
ISIMYCARINX
are uncorrelated
necessarily
whitenoise independent

ECO5 t O IteEe ECO5 eECOIET ECE
varlo sto it E
Isimulatsend
q EdNeon stationery process
Ye Ye t et where Eid Yo of time 0
Y YatE Yotetete
the time seein
Xt I Ye the Ext
a random well
Definition
YaYtEa YotEt

Y is weakly
EL e tee ECT e ECE t ECE e
set wit N10,24
Varle IfVarleye t Varlet
Etty is a stationery
ri tri ÉÉ t T variance
is increasing with the yea’m
between only on j 5th order anclehin
stationary process depends
Enfity A stochasticpreen Eley w ergodic if the randomvariables that are far apartin time

eye r aan o 1 E Yf aid Nco 1 agree because
essentiellyindgundent
independent Suppose X is another RV that is independent f all the Ye’s
Assume X N N 0,1 ZE YetX
allthe Ye’s are

check if Izzy is weakly stationary
1 ECZ ECYeex
EET ELI rage.az
constant Yet Yee t
thx they constant
TIME SERIES MODELS
Varix I weakly stationary
tr adf.esYIaond PROCESS
Mco ri 011
Me ECE e ECOEen DD
Vor Ye Var
It off Iliad a constant
Morina Average
a MAC process y
is an qmy
special case
awn process with
00 weekly stationary
ECYe EMtEttd
me eet d raredEta

Ya Corfeet edgy
E2 thro a varlet
meet ed Ee z
Etc constant c
Gr Yt Yt 2 0 ruo
GlueEtta.ee ne Etat dEes
Lif y effi
set e e E so
Yo I y ItE
i month n stuck return the 2 month sateen would be
Rt 911 y is on
Appliation
Var we can write
we canshow Rely
M t d e dEo
Eat O E continuously compounded
Rem EReEcru 2m
MACH process Var Rta 26
Mt Et I Mt Et l

MACE model Ye u
Ege gaz em
Rely 2MtEEt t.name
in an Mali process with mean a Zee a
Var Rta 62CIt1 262
Ett QstatQEtz t
mean exam PWI
Autorearessivellmon
Disadr MA models longer run correlations require lot of
additional parameters
in an ARID process Waco E
aofYeimtEtwhere dcti
E strength A feedback Treasury internet rates unemployment rates
growthrates gdp industrial production Cpi

程序代写 CS代考 加微信: powcoder QQ: 1823890830 Email: powcoder@163.com