Scribd
Upload
42 views

Assignment 1: Radu Lupu

This document outlines an assignment for a financial econometrics course. Students are asked to: 1) Download stock index price data for at least 5 years from two European or Asian markets and calculate daily returns. Plot the prices and returns and calculate descriptive statistics. 2) Calculate autocorrelations of returns and squared returns up to a lag of 100 days and plot the results. 3) Calculate a GARCH model of volatility over time and plot the standardized returns and their statistics. 4) Calculate returns over different time horizons and compare their distributions. Students must deliver a report by March 30th with description, charts, tables and interpretation of results, packaged with their code and sent to the

Uploaded by

Simona Alexandra Ceburiac
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
42 views

Assignment 1: Radu Lupu

This document outlines an assignment for a financial econometrics course. Students are asked to: 1) Download stock index price data for at least 5 years from two European or Asian markets and calculate daily returns. Plot the prices and returns and calculate descriptive statistics. 2) Calculate autocorrelations of returns and squared returns up to a lag of 100 days and plot the results. 3) Calculate a GARCH model of volatility over time and plot the standardized returns and their statistics. 4) Calculate returns over different time horizons and compare their distributions. Students must deliver a report by March 30th with description, charts, tables and interpretation of results, packaged with their code and sent to the

Uploaded by

Simona Alexandra Ceburiac
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 2

M ASTER IN I NTERNATIONAL F INANCIAL R ISK M ANAGEMENT

C OURSE OF F INANCIAL E CONOMETRICS

Assignment 1

Radu Lupu
March 16, 2020

1 D ATA AND D ESCRIPTIVE S TATS


Download data on a series of stock index prices with a daily frequency, for at least 5 years.
Choose two indices from the European or Asian capital markets.
Use Python or Matlab to solve the following requirements 1

1. Calculate the daily returns as R t +1 = l n(S t +1 /S t ), where S t +1 is the closing price on a


day t +1, S t is the closing price on day t , and l n is the natural logarithm. Plot the closing
prices and returns over.

2. Calculate the mean, standard deviation, skewness and kurtosis of returns. Plot a his-
togram of the returns with the normal distribution imposed as well.

3. Calculate the 1st through 100th lag autocorrelation. Plot the autocorrelations against
the lag order.

4. Calculate the 1st through 100th lag autocorrelation of squared returns. Again plot the
autocorrelations against the lag order.

5. Set σ20 (i.e. the variance of the first observation) equal to the variance of the entire se-
quence of returns (you can square the standard deviation found in 2)). Then calculate
σ2t +1 = 0.94σ2t +0.06R t2 for t = 2, 3, ..., T (the last observation). Plot the sequence of stan-
dard deviations, i.e. plot σt .

1 Solutions in Excel will be made available on the website.

1
6. Compute the standardized returns as z t = R t /σt and calculate the mean, standard de-
viation, skewness and kurtosis of the standardized returns. Compare them with those
found in question 2).

7. Calculate daily, 5-day, 10-day, and 15-day non-overlapping log returns. Calculate the
mean, standard deviation, skewness and kurtosis for all four return horizons. Do the
returns look more normal as the horizon increases?

2 R EPORTING DETAILS
This assignment will be delivered until the 30th of March. It should contain the description
of each step you performed, charts and summary tables, and interpretation of results. All the
files (Excel, Eviews and Matalb for computation and Latex / Word / pdf for report) will be
packed inside an archive file that will have the Surname(s) of the (maximum) 3 persons that
contributed at its realization. This file will be sent to [email protected] by the same
deadline. No other email destinations will be taken into account.

You might also like