ISyE 6402 Midterm Prep
Getting a three variable VAR version from summary(model) output of a VAR(1) version
first matrix: first row are coefficients for Xt1, second row are coefficients for Xt2, and many
others...
Second matrix is Xt-1, i b/c this is a VAR(1) version
last matrix are the constants
eta_t is covariance matrix, direct reproduction
(c) Based at the outfitted model, is there contemporaneous pass-correlation? Is there lagged
cross-correlation? Is there lagged automobile-correlation? Explain.
Contemporaneous cross-correlation is NOT gift if the variance-covariance matrix is a diagonal
matrix
there's lagged correlation if the order p of the VAR(p) model > zero
T/F - Differencing the statistics may not make the series stationary within the presence of
cointegration.
True
Cointegration and lengthy-run equilibrium
See photo
Does cov(x,x) = var(x)?
You betcha
Autocovariance T/F
see photograph
T/F - The AR(1) system is causal if and only if the autoregressive parameter phi is among 0 and
1. However, it's miles always invertible.
FALSE! The absolute cost of phi have to lie b/w -1 and 1
, T/F - A linear method is a special case of the shifting average version.
FALSE - the transferring average is a unique case of a linear technique.
T/F - A guassian time series is usually stationary
fake - guassian strategies can have varying method
T/F 'In autoregressive fashions the current fee of structured variable is prompted through
beyond values of both established and unbiased variables.'
FALSE - there are no analogies of established/independent variables w/ AR models, as there
are w/ regression models
in AR fashions the modern cost of the structured variable is affected by the beyond values of
both structured and independent variables
False - We don't have based and independent variables in AR models like we do in regression
models
how do ACF and PACF differ?
TBD
what in an ACF plot could display non-stationarity?
Slowly reducing lags
what in an ACF and PACF plot might display stationary?
Few lags outside of confidence bands, speedy lowering
can confidence periods be used for importance?
You bet - ought to all be same signal for significance
in a VAR version w/ seasonality for one year, how many seasonality dummy variables will you've
got?
Simply eleven - # of categories - 1
Getting a three variable VAR version from summary(model) output of a VAR(1) version
first matrix: first row are coefficients for Xt1, second row are coefficients for Xt2, and many
others...
Second matrix is Xt-1, i b/c this is a VAR(1) version
last matrix are the constants
eta_t is covariance matrix, direct reproduction
(c) Based at the outfitted model, is there contemporaneous pass-correlation? Is there lagged
cross-correlation? Is there lagged automobile-correlation? Explain.
Contemporaneous cross-correlation is NOT gift if the variance-covariance matrix is a diagonal
matrix
there's lagged correlation if the order p of the VAR(p) model > zero
T/F - Differencing the statistics may not make the series stationary within the presence of
cointegration.
True
Cointegration and lengthy-run equilibrium
See photo
Does cov(x,x) = var(x)?
You betcha
Autocovariance T/F
see photograph
T/F - The AR(1) system is causal if and only if the autoregressive parameter phi is among 0 and
1. However, it's miles always invertible.
FALSE! The absolute cost of phi have to lie b/w -1 and 1
, T/F - A linear method is a special case of the shifting average version.
FALSE - the transferring average is a unique case of a linear technique.
T/F - A guassian time series is usually stationary
fake - guassian strategies can have varying method
T/F 'In autoregressive fashions the current fee of structured variable is prompted through
beyond values of both established and unbiased variables.'
FALSE - there are no analogies of established/independent variables w/ AR models, as there
are w/ regression models
in AR fashions the modern cost of the structured variable is affected by the beyond values of
both structured and independent variables
False - We don't have based and independent variables in AR models like we do in regression
models
how do ACF and PACF differ?
TBD
what in an ACF plot could display non-stationarity?
Slowly reducing lags
what in an ACF and PACF plot might display stationary?
Few lags outside of confidence bands, speedy lowering
can confidence periods be used for importance?
You bet - ought to all be same signal for significance
in a VAR version w/ seasonality for one year, how many seasonality dummy variables will you've
got?
Simply eleven - # of categories - 1
