ISYEl 6402l Midterml Examl (Latestl 2025/l
2026l Update)l Review|l Q/Al |l Gradel A|l
100%l Correctl (Verifiedl Answers)
Q:l Gettingl al 3l variablel VARl modell froml summary(model)l outputl ofl al VAR(1)l model
Answer:
firstl matrix:l firstl rowl arel coefficientsl forl Xt1,l secondl rowl arel coefficientsl forl Xt2,l etc...
secondl matrixl isl Xt-1,l il b/cl thisl isl al VAR(1)l model
lastl matrixl arel thel constants
eta_tl isl covariancel matrix,l directl copy
Q:l (c)l Basedl onl thel fittedl model,l isl therel contemporaneousl cross-correlation?l Isl therel
laggedl cross-correlation?l Isl therel laggedl auto-correlation?l Explain.
Answer:
contemporaneousl cross-correlationl isl NOTl presentl ifl thel variance-covariancel matrixl isl al
diagonall matrix
,therel isl laggedl correlationl ifl thel orderl pl ofl thel VAR(p)l modell >l 0
Q:l T/Fl -l Differencingl thel datal mightl notl makel thel seriesl stationaryl inl thel presencel
ofl cointegration.
Answer:
True
Q:l Cointegrationl andl long-runl equilibrium
Answer:
Q:l Doesl cov(x,x)l =l var(x)?
Answer:
Youl betcha
, Q:l Autocovariancel T/F
Answer:
Q:l T/Fl -l Thel AR(1)l processl isl causall ifl andl onlyl ifl thel autoregressivel parameterl phil
isl betweenl 0l andl 1.l However,l itl isl alwaysl invertible.
Answer:
FALSE!l thel absolutel valuel ofl phil mustl liel b/wl -1l andl 1
Q:l T/Fl -l Al linearl processl isl al speciall casel ofl thel movingl averagel model.
Answer:
FALSEl -l thel movingl averagel isl al speciall casel ofl al linearl process.
Q:l T/Fl -l Al guassianl timel seriesl isl alwaysl stationary
Answer:
falsel -l guassianl processesl canl havel varyingl means
Q:l T/Fl 'Inl autoregressivel modelsl thel currentl valuel ofl dependentl variablel isl influencedl
byl pastl valuesl ofl bothl dependentl andl independentl variables.'
2026l Update)l Review|l Q/Al |l Gradel A|l
100%l Correctl (Verifiedl Answers)
Q:l Gettingl al 3l variablel VARl modell froml summary(model)l outputl ofl al VAR(1)l model
Answer:
firstl matrix:l firstl rowl arel coefficientsl forl Xt1,l secondl rowl arel coefficientsl forl Xt2,l etc...
secondl matrixl isl Xt-1,l il b/cl thisl isl al VAR(1)l model
lastl matrixl arel thel constants
eta_tl isl covariancel matrix,l directl copy
Q:l (c)l Basedl onl thel fittedl model,l isl therel contemporaneousl cross-correlation?l Isl therel
laggedl cross-correlation?l Isl therel laggedl auto-correlation?l Explain.
Answer:
contemporaneousl cross-correlationl isl NOTl presentl ifl thel variance-covariancel matrixl isl al
diagonall matrix
,therel isl laggedl correlationl ifl thel orderl pl ofl thel VAR(p)l modell >l 0
Q:l T/Fl -l Differencingl thel datal mightl notl makel thel seriesl stationaryl inl thel presencel
ofl cointegration.
Answer:
True
Q:l Cointegrationl andl long-runl equilibrium
Answer:
Q:l Doesl cov(x,x)l =l var(x)?
Answer:
Youl betcha
, Q:l Autocovariancel T/F
Answer:
Q:l T/Fl -l Thel AR(1)l processl isl causall ifl andl onlyl ifl thel autoregressivel parameterl phil
isl betweenl 0l andl 1.l However,l itl isl alwaysl invertible.
Answer:
FALSE!l thel absolutel valuel ofl phil mustl liel b/wl -1l andl 1
Q:l T/Fl -l Al linearl processl isl al speciall casel ofl thel movingl averagel model.
Answer:
FALSEl -l thel movingl averagel isl al speciall casel ofl al linearl process.
Q:l T/Fl -l Al guassianl timel seriesl isl alwaysl stationary
Answer:
falsel -l guassianl processesl canl havel varyingl means
Q:l T/Fl 'Inl autoregressivel modelsl thel currentl valuel ofl dependentl variablel isl influencedl
byl pastl valuesl ofl bothl dependentl andl independentl variables.'
