Scribd
Upload
15 views4 pages

REVISION_QUESTIONS_e

The document presents sample questions related to financial modeling, focusing on statistical tests such as the Augmented Dickey-Fuller test and ARMA modeling. It includes tasks for interpreting Stata output, formulating hypotheses, and testing for stationarity and invertibility in time series data. Additionally, it discusses volatility modeling concepts like ARCH and GARCH, and outlines steps for testing the ARCH effect.

Uploaded by

lyeohan Hwa
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
15 views4 pages

REVISION_QUESTIONS_e

The document presents sample questions related to financial modeling, focusing on statistical tests such as the Augmented Dickey-Fuller test and ARMA modeling. It includes tasks for interpreting Stata output, formulating hypotheses, and testing for stationarity and invertibility in time series data. Additionally, it discusses volatility modeling concepts like ARCH and GARCH, and outlines steps for testing the ARCH effect.

Uploaded by

lyeohan Hwa
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 4

Sample Questions 1for financial modelling

Chapter One: Introduction


Question 1
. dfuller logprice, regress trend lags(3)

Augmented Dickey-Fuller test for unit root Number of obs = 16099

Interpolated Dickey-Fuller
Test 1% Critical 5% Critical 10% Critical
Statistic Value Value Value

Z(t) -2.224 -3.960 -3.410 -3.120

MacKinnon approximate p-value for Z(t) = 0.4764

D.logprice Coef. Std. Err. t P>|t| [95% Conf. Interval]

logprice
L1. -.0005913 .0002659 -2.22 0.026 -.0011124 -.0000701
LD. .0300872 .0078826 3.82 0.000 .0146364 .045538
L2D. -.0427619 .0078788 -5.43 0.000 -.0582052 -.0273187
L3D. .0037574 .0078822 0.48 0.634 -.0116926 .0192073
_trend 1.56e-07 7.47e-08 2.09 0.037 9.34e-09 3.02e-07
_cons .0021789 .000839 2.60 0.009 .0005344 .0038235

1) What is the main objective or purpose conveyed by the Stata output presented above
regarding the Dickey-Fuller test and the regression model? (1mark)
2) Provide the complete equation representing the regression model that serves as the
foundation for the Dickey-Fuller test in this output. (2marks)
3) Formulate the null and alternative hypotheses both in plain language and using
statistical notation for the Dickey-Fuller test conducted based on the regression model.
(2marks)
4) Given the test statistic and critical values provided in the output, what decision can be
made regarding the null hypothesis? Provide a justification for your decision and draw
a conclusive statement based on the test results(2marks)
Question 2
Consider the following autocorrelation and partial autocorrelation coefficients estimated using
500 observations for a weakly stationary series, yt:
Use both the Portmanteau and Box–Pierce statistics to test the joint null hypothesis
that the first five autocorrelation coefficients are jointly zero. (5marks)

Chapter Two: ARMA Modelling

Question 1

Question 2

Question Six
Consider the following ARMA (2,3) process
𝑦𝑡 = 0.6𝑦𝑡−1 − 0.3𝑦𝑡−2 + 𝜀𝑡 − 1.4𝜀𝑡−1 + 0.9𝜀𝑡−2 − 0.3𝜀𝑡−3
i) Calculate the mean and variance of 𝑦𝑡
ii) Test for the stationarity and invertibility of this process.
. arima return , ar(1/2) ma(1/3) nolog

ARIMA regression

Sample: 04jan1950 - 31dec2013 Number of obs = 16102


Wald chi2(5) = 20058.33
Log likelihood = 51732.56 Prob > chi2 = 0.0000

OPG
return Coef. Std. Err. z P>|z| [95% Conf. Interval]

return
_cons .0002924 .0000811 3.60 0.000 .0001334 .0004514

ARMA
ar
L1. -1.58899 .0435536 -36.48 0.000 -1.674354 -1.503627
L2. -.6334204 .0435005 -14.56 0.000 -.7186797 -.548161

ma
L1. 1.620282 .043896 36.91 0.000 1.534247 1.706316
L2. .6477836 .046755 13.85 0.000 .5561454 .7394218
L3. -.0202577 .0044904 -4.51 0.000 -.0290587 -.0114567

/sigma .0097378 .0000156 624.86 0.000 .0097072 .0097683

Note: The test of the variance against zero is one sided, and the two-sided
confidence interval is truncated at zero.

1) Clearly state the statistical process being estimated in the Stata output. (1mark)
2) Write the estimated equation of the process, rounding coefficients to two decimal
places. (2 marks)
3) Calculate and provide the unconditional mean of the process. (1.5 marks)
4) Develop characteristic equations for both the AR and MA components of the process.
Present the equations concisely. (2 marks)
5) Test for stationarity and invertibility, discussing the criteria and concluding on the
properties of the process. (4 marks)

Chapter 3: Modelling Volatility

1) What are some stylized facts of financial time series data that volatility models aim to
capture?
2) Explain the fundamental concept behind ARCH (Autoregressive Conditional
Heteroskedasticity) models.
3) Differentiate between ARCH and GARCH (Generalized Autoregressive Conditional
Heteroskedasticity) models. What additional feature does GARCH introduce?
4) Contrast implied volatility and historical volatility.
5) Describe the procedural steps involved in testing for the ARCH (Autoregressive
Conditional Heteroskedasticity) effect in a time series
6) Demonstrate that the GARCH (1,1) process can be viewed as an ARMA(1,1) process
for the conditional variance
7) In the context of volatility modelling, it is widely acknowledged that GARCH is often
deemed more parsimonious than ARCH. Demonstrate this by illustrating that the
GARCH(1,1) model can be equivalently expressed as an ARCH model of infinite order

You might also like