You are on page 1of 2

SKEMA Juliana CAICEDO-LLANO

Master FMI julicaicedo@gmail.com


Paris-Sophia 2023-2024

Introduction to Financial Econometrics


SKEMA
Document 1

1 Simple Linear Regression


Document 1 - CAPM
The following figures represent prices of stocks and levels of equity indexes or its returns as well as results from simple
linear regressions. Analyze the figures and answer to the questions.

Figure 1 – Equity prices Figure 2 – Market Index


120

BNP
CAR
DAN
RNO

6000
100

5000
80
ACTIONS_FR

CAC
60

4000
40

3000
20

2004 2006 2008 2010 2012

Index
2004 2006 2008 2010 2012

Figure 3 – Daily returns Figure 4 – Equity returns vs Index returns

rentabilité: Action BNP − Indice de marché


800


0.10



● ●


● ●

600

0.05

●● ● ●
● ●● ●

●●● ●●
● ●
● ●● ●●● ● ● ●
●●● ● ● ●
● ● ● ● ●●● ●●●● ●● ● ● ●

● ● ● ●● ● ●●●●● ● ●● ●●●● ●
● ●● ●● ●●●●●●
● ● ●●
●● ●●●●●●●● ●●●●●
Frequency


●●
●●●●
● ●●●
● ●
● ● ●●●●●● ●
●●●

●●●●


●●


●●


●●

●●
●●
●●

● ●












●●

●● ●●
●●●●
●●●● ● ● ●
● ● ●
● ● ●● ●● ●●
●●
●●
●●

●●●
● ●
●●
●●


●●

●●

●●



●●
●●
● ●
●●●●●●
●●● ●●
●●●●● ●●● ● ●●


●●●●
●●● ● ● ● ●● ● ●● ●
CAC

● ●●●
● ●● ●●
●●●●●
● ●
●● ● ● ● ●
●●●● ●●●
● ● ●●
●●
●●
●●

●●


●●●

●●
●●

●●


●●
●●
●●
●●
●● ●●●●●● ● ●●
400

● ●●
●●
●●
●●
● ● ●●
●●●
● ●● ●
●● ● ●●●● ●●● ●●
●●●
●●
● ●
●●
●●
●●●

●●
●●●

●●
●●●

●●

●●
●●●
●● ●
●●●● ●●
● ● ●
0.00

● ●●
● ●

●●


●●

●●

●●

●●


●●
●●

●●

●●●

●●

●●●

●●●
●●
●●
● ●

●●●
●● ●
● ●● ●● ●●●●● ●●●● ●
●●
●●●

●●

●●

●●

●●



●●●

●●
●●

●●



●●


●●

●●
●●

●●●
●●●

●●●

●●●●●
●●●● ●
● ●
● ●●●
● ●

●●

●●

●●

●●
●●
●●



●●
●●




●●



●●



●●


●●

●●

●●


●●

●●


● ●●●●●●
● ●

●●● ●
● ● ●
● ●● ●●● ●● ●
● ● ●
●● ●
● ●
●●

●●●●

● ●
●●●
●● ●
●● ●●
● ●●● ●● ● ●
● ●
● ●● ●● ●●●●●

●● ●

●●●●●
●●●
●●


●●


●●



●●


●●

●●●





●●


●●



●●

●●


●●
●●●
●●

●●●


● ●●●●
●●● ● ●●
● ● ●●
● ●● ●●


● ●
●●●●
●●●


●●



●●


●●


●●●
●●
●●●
●●●●


●●
● ●●●●


●●●● ●●●●●●● ●●● ●●●●

●● ●●

●●●
●●
●●


●●

●●
●●
●●
●●
●●

● ●




● ●
●●●●●●●
● ●●
● ● ● ● ●
● ●●
●●●●●
● ●
●●

● ●


●●
● ●●●●●
●●
●●
●●
●●
●●
●● ●
●●●
●● ● ●● ● ●
●● ●●●● ●
●●●
●●●● ●● ●





●●
● ●

●●
●●●










●●●


● ●●● ●
● ●●●●●● ● ●
●● ●●
●●●●● ●●●●●●●
● ●●●● ● ●

● ●●●●●●●
●●
● ●●
●●●●●
●● ●
●● ●

●●
●●●●

● ●●●●●
● ● ●● ● ●
● ● ● ● ●● ●
● ● ● ● ● ●●● ● ●●● ● ●
● ●● ●● ●●
●●● ● ●
200

● ●
−0.05

●● ● ●● ●
● ● ● ● ●

● ●
● ● ●● ●●

● ●
● ●

−0.10


0

−0.10 −0.05 0.00 0.05 0.10 −0.2 −0.1 0.0 0.1 0.2

BNP

Questions 1 - CAPM
1. Using figures 1 and 3, how is the probability distribution of the series ?
2. What is the difference between the series in figures 1 and 2 and those in figures 3 and 4 ? Why using
this transformation ?
3. Using figure 4 (Excess returns of stock BNP and CAC index), do you think it is possible to test the
CAPM ? These graphs use daily series, do you think the same results will hold with monthly series ?

1
4. How do you interpret the coefficients from the estimations in figure 5 ?
5. The parameters of the CAPM are estimated for Danone (DAN), do they validate the CAPM model ?
Which hypothesis test are useful to validate it ?
6. Is it possible to classify the different stocks based on their beta coefficients ? All these regression have
the same explanatory power ?
7. Figure 6 presents the results of regressions of Renault’s returns over the market index for two different
sub-samples, what can we conclude from these results ?
8. If the forecast for next month market excess return is 1 %, which are the forecasts and confidence
intervals for each of the stocks ?

Figure 5 – Equity Regressions Figure 6 – Regression - Renault

BNP CAR DAN RNO Model 1 Model 2


(Alpha) (Intercept) 0.00 −0.01 0.00 −0.01 (Intercept) −0.01 0.00
(0.01) (0.01) (0.00) (0.01) (0.01) (0.01)
(Xt (daily
rCACm 1.34∗∗∗ 0.78∗∗∗ 0.58∗∗∗ 2.23∗∗∗ rCACm[1 :54, ] 1.85∗∗∗
returns))
(0.11) (0.11) (0.08) (0.14) (0.25)
rCACm[55 :108, ] 2.36∗∗∗
R2 0.57 0.30 0.31 0.70
(0.19)
Adj. R2 0.56 0.30 0.31 0.70
Num. obs. 108 108 108 108 R2 0.52 0.75
Adj. R2 0.51 0.75
*** p
< 0.01, ** p < 0.05, * p < 0.1 Num. obs. 54 54
Each coefficient has an associated standard error
*** p< 0.01, ** p < 0.05, * p < 0.1
—> We cant compute that because the error term is
normally distributed Figure 6
With SLR, you do not use the adjusted R^2 1st perdiod = Model 1 = 54 first obs
Documents 2 - Hedge Ratio 2nd period = Model 2 = 54 last obs
The following figures present the S&P 500 index and its shortest maturity futures contract, the linear regressions are
used to estimate the optimal hedge ratio. The variables used for regressions are in logarithms or returns. Analyze these
figures and answer to the questions.

Figure 7 – SP500 et Future SP500 Figure 8 – Ratio optimal

rSP lSP
(Intercept) 0.00 0.11
SPX Index
SP1 Index (0.00) (0.18)
1600

rFut 0.12
1400

(0.13)
lFut 0.98∗∗∗
Ct[, Uni]

1200

(0.03)
1000

R2 0.01 0.96
Adj. R2 -0.00 0.96
800

Num. obs. 65 66
600

0 1000 2000 3000


RMSE 0.04 0.03
∗∗∗ p < 0.001, ∗∗ p < 0.01, ∗ p < 0.05

Questions 2 - Hedge Ratio


1. What is optimal hedge ratio using returns ? Using logs ?
2. What can you say about the significance of the coefficients ? Which test of hypothesis have been performed ?
3. What can you say about the significance of the determination coefficient ?
4. How can you interpret the results from these two regressions ?

You might also like