ECON334 Financial Econometrics JUNE 2017
Hello, dear friend, you can consult us at any time if you have any questions, add WeChat: daixieit
FORMAL EXAMINATION PERIOD: SESSION 1, JUNE 2017
ECON334
Financial Econometrics
Part A-Multiple Choice Questions (25 in total,1 mark each)
Question 1
Consider the following regression model: y,=β+β,x,+β₂z,+u,which we have estimated using a sample of 200 observations. Suppose we conduct White's test for heteroscedasticity in the regression residuals, at significance level of 5%, using the auxiliary regression:
The R² of the auxiliary regression is 0.051 and the test statistic is 10.2. The appropriate critical value with which we should compare the test statistic is:
11.07
12.83
12.59
1.96
1.64
Question 2
Which of the following could be used to test for autocorrelation in regression residuals up to fourth order?
(a) Durbin Watson test only
(b) LM (Breusch Godfrey) test only
d(c) Both (A) and (B) above
(e) All of the above
Question 3
If we have detected autocorrelation in the residuals of a linear regression then which of the following are plausible approaches to dealing with the problem?
(i) add lagged values of the dependent variable to the right hand side of the equation
(ii) use dummy variables to remove outliers in the data
(iii) use robust standard errors together with the coefficient estimates (iv) use logs of the dependent variable and explanatory variables
(a) (b) (c) (d) (e)
all of (i),(ii),(iil) and (iv)
(i) and (iii) only
(i) only
Question 4
The ACF and PACF of a time series are shown in the following graph. They indicate that the model for the time series is best characterised as:
Lags
(a) ARMA(1,1)
(b) MA(1)
(c) MA(2)
(d) AR(1)
(e) AR(2)
Question 5
Consider the following process:
y₁=0.55+0.4u-2+0.2u- 1+,
What is the optimal forecast of y+2, given that all information up to and including time t is available; in particular that u,=0.3, M,=-0.6 and u-z=0.4.
0.95
(b)
(c)
(d)
(e) None of the above
Question 6
The random variable y, follows the MA(2) process given by: y,=u,-0.4u,-1-0.2u,-2 where u, is a white noise process with a variance of 0.5. The population variance of y, is:
(a)
())
(e(d))
0.6
1.2
0.2
0.0
1.0
Question 7
The random variable y, follows the ARMA(2,1) process given by:
y₁=10+0.8y,-+0.1y,-2+1,-0.4u- where u,is a white noise process. The unconditional mean of y, is:
(c(b))
(e(d))
10.0
100
0.0
5.26
1.0
Question 8
For the process y,=1.3y--0.4y,2+u, where μ, is a white noise process, consider the following statements about the stationarity of y,
(i) the relevant characteristic polynomial is 1- 1.3z+0.4z²
(ii) the roots of the relevant characteristic polynomial are 2.00 and 1.25
(i) the relevant characteristic polynomial is 1+1.3z-0.4z² with roots equal to 3.89
and-0.64
(iv) y, is a stationary process
(v)y, is a non-stationary process
Which of the above statements are true concerning the stationarity of y,?
(a) (i) and (v) only
(b) (i) and(iv) only
(c) (ii) and (iv) only
(d) (v) only
(e) (i),(ii) and(iv) only
Question 9
An AR(2) model can be written using lag operator notation as
中(L)y,=c+u,
where u, is a white noise process and where
φ(L)=(1-4L-φ₂L²)
If y, is a stationary process then which of the following statements are true?
(i) The ACF and PACF of y, will both exhibit exponential decay
(ii) the roots of the characteristic polynomial 1-hz-z² lie inside the unit circle (i) There exists an alternative representation of y, as an infinite length MA process (iv) The inverse of φ(L) exists and is represented by φ'(L)=(1-φL-,L²)¹
(a) (iii) and (iv) only
(b) (i) and (iv) only
(c) (i) only
(d) (i),(iii) and (iv) only
(e) none of the statements are true
Question 10
A researcher is interested in forecasting the stock price index in a certain country. The observed stock price index values from 2012 to 2016 are shown in the table along
with their forecast values from some forecasting model.
Year |
Observed Value |
Forecast value |
2012 |
101 |
100.5 |
2013 |
103 |
102.4 |
2014 |
104 |
103.2 |
2015 |
107 |
106.0 |
2016 |
111 |
111.0 |
The mean squared forecast error is:
(a)
(b)
(c)
(e(d))
0.58
0.98
0.45
1.95
0.73
Question 11
A stationary autoregressive process of order two (i.e. an AR(2)) would have the following characteristics:
(a) a decaying ACF and PACF
(b) an ACF and a PACF with two significant spikes
(c) an ACF with one significant spike and an PACF with two significant spikes
(d) a decaying PACF, and an ACF with two significant spikes
(e) a decaying ACF, and a PACF with two significant spikes
Question 12
The process y, is a zero-mean white noise process if
(a) E(y,)=0 for all t,only
(b) var(y,)=σ² for all t, only
(c) cov(y,,y,-s)=0 for all s≠0,only
(d) Conditions (a) and (b) hold only
(e) Conditions (a),(b) and (c) all hold
Question 13
Which of the following statements are likely to be true in relation to the behaviour of daily returns from shares traded on a major stock exchange?
(i) returns cannot usually be forecast with any accuracy because the time series of stock prices resembles a random walk.
(ii) ARCH and GARCH can be useful for modelling common patterns in the variance
of returns such as volatility clustering
(iii) share prices usually fluctuate around a deterministic time trend
(iv) there is no pattern in either share returns or in the variance of returns.
(a) (i) only
(b) (iii) only
(c) (iv) only
(d) (i) and (ii) only
(e) (i) and (ii) and(iii)
Question 14
Consider the GARCH(1,1) model for conditional variance given by
where u, is the residual from a regression equation (i.e. the mean equation) at time t.We have the following parameter estimates w =0.06,α=0.3,β=0.2,and we are given
=0.4 and σo=0.3.Calculate the conditional variance at time t=1 (i.e.σ²). The correct answer is given by:
(a)
(b)
(c)
(d)
(e)
0.126
0.240
0.168
0.198
0.060
Question 15
Consider the following statements about ARCH and GARCH models:
(i) They are used to model the conditional variance of a time series (ii) They model volatility clustering in the data
(i) They can only be applied to stationary time series (iv) They are estimated by OLS
(v) They can only be applied to non-stationary time series
Which of those statements is correct:
(a) (i) and(ii) only
(b) (i),(i) and(iii) only
(c) (i),(ii) and (v) only
de (i),(ii) and (iv) only
Question 16
Consider an MA(1)-GARCH(1,1) model of returns, given by:
y₁=c+θu- 1+u,
What are the conditions that need to be satisfied to ensure that u, has a well-defined variance?
(a) αo>0,α₁ ≥0,β≥0
(b) αo+α₁+β<1
(c) α₁+β<1
(d) Conditions (a) and (b)
(e) Conditions (a) and (c)
Question 17
The following is the GJR extension of the GARCH model for equity returns:
r=δ+u
where
I,-=1 if u-<0 and I,-=0 otherwise.
u, is white noise and r; is the return on a stock market.
We are given the fitted equation:
Given and a positive residual in the last period, ti =0.5, the conditional variance is:
(a)
(b)
(c)
(d)
(e)
1.625
1.500
1.000
1.200
1.475
Question 18
Suppose we have a sample of the weekly log price (level) of the S&P/ASX200 share price index, which we denote by In P. We have considered a number of possible ARMA(p,q)
models for the process which generates the time series of weekly log differences of the
index, i.e. for △ln P. The best model for weekly log differences according to the AlC criteria
is an MA(1) model, and the fitted regression equation is
△lnP=0.004-0.05
where △ln P means the change in the log of the (stock) price level, and u, is the regression residual. Which of the following statements are likely to be correct about the true process for InP?
(i) In P is approximately a random walk with drift
(ii) InP is an MA(1) process
(ii) In P, is stationary around a linear time trend
(iv) In P, is an l(1) process
(v) InP is an |(2) process
(a) (i) and(ii) and (iv) only
(b) (i) only
(c) (v) only
de (i) and (iv) only
Question 19
The data generating model for first differences of y, is given as follows:
Ay,=Ay,-₁+u, where u, is a white noise process.
Which of the following statements is NOT true?
(a) y, is non-stationary
(b) Ay,is stationary
(c) y, is an l(2) process
(d) Ay, is a random walk
(e) Ay, is an I(1) process
Question 20
Which condition is necessary for log(σ;) to have the same response to positive and negative shocks in the EGARCH model below?
(a)
(b)
(c)
(d)
(e)
w=0
α=0
α>0
y=0
y>0
Question 21
Consider the bi-variate vector autoregression (VAR) model below:
y₁=αo+α₁y- 1+α₂X₁- 1+α3y₁-2+a₄X,-2+V
x,=B₀+βy₁- 1+β₂x- 1+β3y₁-2+β₄x-2+V₂
We say that x, does NOT Granger cause y, if
(a) lagged values of x, do not help to predict y, beyond the information already contained in lagged values of y;
(b) lagged values of y, do not help to predict x, beyond the information already contained in lagged values of x,
(c) E[y: |y₁-1,y-2]=E[y₁ |y₁-,y₁-2,X,-1,x-2]
(d) (a) and (b) and (c) (e) (a) and (c) only
Question 22
Suppose the variables Y, and X, are both I(1) variables and are NOT cointegrated with each other. Which of the following statements is true?
(i) in a regression of Y, on X,, the regression residual is stationary
(i) a regression of Y, on X, is a spurious regression
(iii) there exists a linear combination of Y, and X, which is stationary
(iv) in a regression of Y, on X,, the regression residual is non-stationary
(a) only statement (i) is true
(b) statements (i) and (iii) are true
(c) statements (ii) and (iv) are true
(d) statements (i),(ii) and (iii) are true
(e) None of the statements are true
Question 23
ARCH and GARCH models are estimated using the:
(a) OLS estimation method
(b) OLS estimation method with an adjustment for heteroscedasticity
(c) method of instrumental variables
(d) method of maximum likelihood
(e) none of the above
Question 24
A researcher would like to run an augmented Dickey-Fuller (ADF) test on the variable y.
What is the regression that would be estimated and what is the null hypothesis (Ho)of the test?
(a) and H 。:y =0, respectively
(b) , and H 。:y=0, respectively
(c) and H 。:y =0, respectively
(d) and H 。:y =1, respectively
(e) ,and H 。: =0,respectively
Question 25
Robust standard errors should be used for reliable inference when the residuals from an estimated regression are found to be:
(a) autocorrelated and/or heteroscedastic
(b) indicative of an incorrect functional form
(c) those from a cointegrating regression
(d) those from a spurious regression
(e) uncorrelated and homoscedastic
Part B- There are TWO Questions. Each question has parts to it and each question is worth 10 marks in total. Answer BOTH questions in the examination booklet provided.
Question 1
(a) Suppose that a researcher has data on the log stock price indices of Korea (k,),
Japan(ji)and Singapore(s,). Assume that k,, j, and s, are non-stationary time
series.
(i) The researcher would like to test for cointegration between the Korean and Japanese stock markets. Describe how the researcher would perform this test. Be sure to state the null and alternative hypothesis, and what you
conclude if you reject the null hypothesis (For your information, the ADF 5%
critical value is -2.891 and the Engle-Granger 5% critical value is -3.398).(3 marks)
(ii) Suppose the researcher estimated the following regression:
k,=B₀+β,s,+u,
The autocorrelation function of the estimated residuals (t,) was
approximately one for all lags. Can reliable inferences be drawn from this regression? Explain your answer.(2 marks)
(b) You estimated the following AR(2) model for some time series:
y₁=0.5+1.3y--0.4y,-2+u,
(i) ls the AR(2) model stationary ? Justify your answer.(2 marks)
(ii) Obtain an expression for the two-step ahead forecast f or the series,
conditional on time t information i.e. find an expression for E(y₂ 1Ω,) where 2, is the information set at time t which includes observations on the y series dated t and earlier.(2 marks)
(i) Using the expression you derived in part (ii), what is the two-step ahead forecast, given that y,=0.5 and yʊ=1.0.(1 mark)
Question 2
A researcher has data on the daily percentage return on Qantas stock from 1 January 2008 to 4 December 2016, a total of 2,153 observations. The return series is denoted r qan.The researcher estimated the regression: r qan =β+1,,and saved the estimated residuals u. The estimate of βo was 0.009 with a t-statistic of 0.18
(a) Do you agree with the statement that the average daily return on Qantas stock
over the sample is small and positive but not significantly different from zero? Justify your answer.(1 mark)
(b) The researcher decided to perform an LM test for third-order ARCH effects. Write down the null and alternative hypothesis for this test, and the auxiliary regression for this test. The auxiliary regression of the test produced an R² of 0.0133.What do you conclude?(2 marks)
The researcher estimated the following ARCH model
r qan=β+u,
o²=ao+am₁ +am²2+α₃²3
Eviews gave the following results:
Dependent Variable:R QAN
|
Coefficient |
Std. Error |
z-Statistic |
Prob. |
居 |
0.057 |
0.046 |
1.238 |
0.2158 |
Variance Equation |
||||
αo α₁ α₂ α₃ |
3.632 0.173 0.093 0.123 |
0.125 0.019 0.020 0.009 |
29.01 9.33 4.72 13.66 |
0.00 0.00 0.00 .00 |
Log-Likelihood =-4849.47
(c) Is the conditional variance always positive in this model? Justify your answer. What does the long-run forecast of the conditional variance converge to in t
2023-06-17