Agenda
Linear regressions
Simple linear model
Ordinary Least Squares
Estimations
Inferences
Applications
1
Financial Econometrics
Application of statistical techniques to problems in finance
Estimating and testing relationships between variables as
suggested by theories in finance, e.g. CAPM or APT
Modeling and forecasting asset returns
Evaluating the performance of portfolio management
Examining determinants of firms’ financial decisions
Steps involved in the formulation of econometric models
Economic or Financial Theory (Previous Studies)
Formulation of an Estimable Theoretical Model
Collection of Data
Model Estimation
Is the Model Statistically Adequate?
No
Reformulate Model
Yes
Interpret Model
Use for Analysis
2
Simple Regression: An Example
Suppose that we have the following data on the excess returns
on a fund manager’s portfolio (“fund XXX”) together with the excess returns on a market index:
Year, t
Excess return
= rXXX,t – rft
17.8
39.0
12.8
24.2
17.2
1
2
3
4
5
Excess return on market index
= rmt - rft
13.7
23.2
6.9
16.8
12.3
We have some intuition that the beta on this fund is positive,
and we therefore want to find whether there appears to be a relationship between x and y given the data that we have.
Graph (Scatter Diagram)
Excess return on fund XXX
45
40
35
30
25
20
15
10
5
0
0
5
10
15
20
25
Excess return on market portfolio
3
Finding a Line of Best Fit
We can use the general equation for a straight line,
y=α+βx to get the line that best “fits” the data.
However, this equation (y=α+βx) is completely deterministic.
Is this realistic? No. So what we do is to add a random
disturbance term, u into the equation. yt = + xt + ut where t = 1,2,3,4,5
Ordinary Least Squares
The most common method used to fit a line to the data is
known as OLS (ordinary least squares).
What we actually do is take each distance and square it and
minimise the residual sum of the squares (hence least squares). Tightening up the notation, let
yt denote the actual data point t denote the fitted value from the regression line
ˆ
ˆ ut denote the residual, yt - yt
ˆ
yt
4
Determining the Regression Coefficients
Choose and so that the (vertical) distances from the data
points to the fitted lines are minimised (so that the line fits the data as closely as possible): y yt
ˆ
ut
ˆ
yt
How OLS Works
5
ˆ
ˆ
ˆ
ˆ
ˆ
ˆ
So min. u 12 u 22 u 32 u 42 u 52 , or minimise u t2 . This is
known as the residual sum of squares.
t 1
But what was ut ? It was the difference between the actual
ˆ
point and the line, yt - yt .
ˆ
ˆ
u
with respect to
So minimising
2 t is equivalent to minimising and .
y
ˆ
yt
2
t
5
Deriving the OLS Estimator
ˆ
ˆ
ˆ
y t x t , so let L
ˆ
ˆ ˆ
( y t y t ) 2 ( y t xt ) 2 t i
Want to minimise L with respect to (w.r.t.) and
, so
differentiate L w.r.t. and
L
ˆ
ˆ
2 ( yt xt ) 0
ˆ
t (1)
L
ˆ
ˆ
2 xt ( yt xt ) 0
ˆ
t
(2)
`
From (1),
ˆ ˆ
ˆ ˆ
( y t x t ) 0 y t T x t 0 t We know y t T y and
x
t
.
Tx
Deriving the OLS Estimator (cont’d)
ˆ
So we can write T y T T x 0 or y ˆ ˆ x 0
ˆ
From (2),
ˆ
x t ( y t ˆ x t ) 0
(3)
(4)
t
ˆ
ˆ
From (3), y x
(5)
Substitute into (4) for from (5),
x t ( y t y ˆ x ˆ x t ) 0 t 2
x t y t y x t ˆ x x t ˆ x t 0 t 2
x t y t T y x ˆ T x 2 ˆ x t 0 t 6
Deriving the OLS Estimator (cont’d)
Rearranging for ,
ˆ ( T x
2
x t2 ) T y x x t y t
So