Reading 21 Stationary Time Series

You might also like

Download as pdf or txt
Download as pdf or txt
You are on page 1of 4

Question #1 of 9 Question ID: 1268271

All of the following statements represent conditions for a time series to be covariance

stationary except:

A) the time series volatility around its mean does not change over time.

B) the covariance of the time series with leading or lagged values of itself is constant.

C) the standard deviation of the time series is in nite, but constant over time.

D) the expected value of the time series is constant over time.

Question #2 of 9 Question ID: 1268274

When examining the properties of a general finite-order process of order q [MA(q)], which of

the following statements is correct when comparing the MA(q) process to the MA(1) process?

I. Both the MA(1) and MA(q) processes exhibits autocorrelation cutoff after the first
lagged error term.

II. The MA(q) process is a subset of the first-order moving average [MA(1)].

A) I only.

B) II only.

C) Neither I nor II.

D) Both I and II.

Question #3 of 9 Question ID: 1268272

Regarding a white noise process, which of the following characteristics should not be
included in the dynamic structure of white noise?

A) Events in a white noise process exhibit no correlation between the past and present.

Both conditional and unconditional means and variances are the same for an
B)
independent white noise process.
The unconditional mean and variance must be constant for any covariance
C)
stationary process.

D) The process must be serially independent and normally distributed.

Question #4 of 9 Question ID: 1268276

A high Q-statistic describes a fund with:

A) large positive autocorrelation or large negative autocorrelation.

B) large negative autocorrelation, only.

C) no autocorrelation.

D) large positive autocorrelation, only.

Question #5 of 9 Question ID: 1268277

If a forecaster is using a time series model and notices periodic spikes in autocorrelations as

they gradually decay, this is most likely a sign of:

A) autoregressive conditional heteroskedasticity (ARCH).

B) seasonality in the data.

C) rst di erencing lag operators.

D) a structural shift in the time series.

Question #6 of 9 Question ID: 1268270

Which of the following terms is most likely associated with the degree of correlation and
interdependency between data points in a time series?

A) Autocorrelation function.

B) Covariance stationary.

C) Autocovariance function.

D) Autoregression.
Question #7 of 9 Question ID: 1268273

Regarding the properties of a first-order moving average [MA(1)] process, which of the
following statements is most likely correct?

A) The MA(1) process has a mean of zero and a dynamic variance.

An MA(1) process is a nonlinear regression of the current values of a time series


B)
against the previous unobserved error terms.

The MA(1) process is considered to be rst-order because it only has one lagged
C)
error term.

For any value beyond the rst lagged error term, the autocorrelation will be one in
D)
an MA(1) process.

Question #8 of 9 Question ID: 1268278

If an analyst notices that autocorrelations decay gradually over time, which time series
process should the analyst most likely rule out when attempting to forecast data?

A) General pth order autoregressive [AR(p)] process.

B) First-order autoregressive [AR(1)] process.

C) Moving average [MA(1)] process.

D) Autoregressive moving average (ARMA) process.

Question #9 of 9 Question ID: 1268275

Zack Snyder is comparing and contrasting the Box-Pierce Q-statistic with the Ljung-Box Q-
statistic. Which of the following statements should he identify as being incorrect regarding
these Q-statistics?

I. The Box-Pierce Q-statistic reflects the absolute magnitudes of the correlations,


because it sums the squared autocorrelations.

II. For large sample sizes, weights for both the Box-Pierce Q-statistic and the Ljung-Box
Q-statistic are roughly equal.
A) II only.

B) I only.

C) Neither I nor II.

D) Both I and II.

You might also like