MFE5150Financial Data Analysis
2015-16 Second Term
Solution of Assignment 2
March 23, 2016
prepared by Bojun LU
Problem 1.
Solution. We know that the bivariate normal distribution of (x1 , x2 ) is as follows:
2
x1 2x1 x2 + x22
1
exp
,
f (x1 , x2 ) = p
2

MFE 5150 Assignment 2
Due Date: 6pm, Wednesday, Mar 16, 2016
Please submit a soft-copy of your solutions to the moodle by the due date.
1. Suppose (X1 , X2 ) has the bivariate normal distribution with density function
2
x1 2x1 x2 + x22
1
f (x1 , x2 ) =

MFE 5150 Assignment 1
Due Date: 5pm, Friday, Feb 26, 2016
1. Consider the linear regression model
yi = 0 + 1 xi1 + 2 xi2 + 3 xi3 + i ,
1 i n.
Using the F -test of a general linear hypothesis, show how to test the following hypotheses on the regression coe

Financial Data Analysis
Time Series
Week 11-13
1 / 19
Univariate time series
I
A time series cfw_x1 , x2 , x3 , . . . is said to be weakly stationary (or
covariance stationary) if Ext and Cov(xt , xt+k ) do not depend on
t 1.
I
For a weakly stationary seq

MFE5150Financial Data Analysis
2015-16 Second Term
Solution of Assignment 1
March 1, 2016
prepared by Bojun LU
Problem 1.
Solution.
(a) We first transform the initial given linear regression model
yi = 0 + 1 xi1 + 2 xi2 + 3 xi3 + i , 1 i n
(1)
to the foll

Financial Data Analysis
Nonparametric regression
Sect. 7.1-7.4
Week 9-10
1 / 18
Nonparametric regression
I
Given a set of observations cfw_xi , yi , i = 1, . . . , n, we want to find
the relation between x and y.
I
Weierstrass approximation theorem: every