Exercises Chapter 1 Estimation Theory Advanced Econometrics - HEC Lausanne Christophe Hurlin University of Orléans November 2013 Christophe Hurlin (University of Orléans) Advanced Econometrics - HEC Lausanne November 2013 1 / 68 Exercise 1 Rayleigh distribution Christophe Hurlin (University of Orléans) Advanced Econometrics - HEC Lausanne November 2013 2 / 68 Problem We consider two continuous independent random variables X and Y normally distributed with N 0, σ2 . The transformed variable R de…ned by: p R = X2 + Y2 has a Rayleigh distribution with a parameter σ2 : R Rayleigh σ2 with a pdf fR r ; σ2 de…ned by: r2 r exp σ2 2σ2 r π E (R ) = σ V (R ) = 2 fR r ; σ 2 = Christophe Hurlin (University of Orléans) 8r 2 [0, +∞[ 4 Advanced Econometrics - HEC Lausanne π 2 σ2 November 2013 3 / 68 Problem (cont’d) We consider an i.i.d. sample fR1 , R2 , .., RN g and an estimator (MLE) of σ2 de…ned by 1 N 2 b2 = σ Ri 2N i∑ =1 b2 is an unbiased estimator of σ2 . Question 1: Show that σ Christophe Hurlin (University of Orléans) Advanced Econometrics - HEC Lausanne November 2013 4 / 68 Solution: We have: b2 = σ 1 2N N ∑ Ri2 i =1 b2 . We want to calculate E σ Since the sample fR1 , R2 , .., RN g is i.i.d., we have: ! 1 N 2 1 N 2 b =E E σ R = ∑ E Ri2 i 2N i∑ 2N =1 i =1 Christophe Hurlin (University of Orléans) Advanced Econometrics - HEC Lausanne November 2013 5 / 68 Solution (cont’d): b E σ 2 =E 1 2N N ∑ i =1 Ri2 ! = 1 2N N ∑E Ri2 π σ2 i =1 We know that: E (Ri ) = σ r π 2 V (Ri ) = 4 2 So, we have: E Ri2 Christophe Hurlin (University of Orléans) = V (Ri ) + E (Ri )2 π 4 π σ2 + σ2 = 2 2 2 = 2σ Advanced Econometrics - HEC Lausanne November 2013 6 / 68 Solution (cont’d): b2 E σ = = = 1 2N 1 2N N ∑E Ri2 i =1 N ∑ 2σ2 i =1 N2σ2 2N So, we have: b2 is unbiased. The estimator σ Christophe Hurlin (University of Orléans) b2 = σ2 E σ Advanced Econometrics - HEC Lausanne November 2013 7 / 68 Remark: Sometimes, the Rayleigh distribution is parametrized by σ. But, b2 is unbiased than to show that σ b is unbiased... it is easier to show that σ v u u 1 N b=t Ri2 σ 2N i∑ =1 Then to study the bias, we have to calculate: 0v 1 u u 1 N b) = E @t E (σ Ri2 A 2N i∑ =1 ??? since for a nonlinear function g (.) , E (g (x )) 6= g (E (x )) ... The only solution is to compute the integral R∞ b ) = 0 x fσb x; σ2 dx E (σ Christophe Hurlin (University of Orléans) Advanced Econometrics - HEC Lausanne November 2013 8 / 68 Problem (cont’d) b2 is a (weakly) consistent estimator of σ2 . We Question 2: Show that σ admit that the raw moments of R are de…ned by: E Rik k = σk 2 2 Γ 1 + k 2 k2N and where Γ (.) denotes the gamma function with: Γ (x ) = Z∞ tx 1 exp ( t ) dt. 1) ! for x 2 N 0 and Γ (x ) = (x Christophe Hurlin (University of Orléans) Advanced Econometrics - HEC Lausanne November 2013 9 / 68 Solution: b2 . Since the sample fR1 , R2 , .., RN g is i.i.d., we First, calculate V σ have: ! N 1 N 1 2 b2 = V R = V Ri2 V σ i 2 ∑ 2N i∑ 4N =1 i =1 What is the value of V Ri2 ? V Ri2 = E Ri4 E Ri2 2 We shown that E Ri2 = 2σ2 Christophe Hurlin (University of Orléans) Advanced Econometrics - HEC Lausanne November 2013 10 / 68 Solution (cont’d): V Ri2 = E Ri4 E Ri2 2 = E Ri4 4σ4 What is the value of E Ri4 ? For any k 2 N : E Rik k = σk 2 2 Γ 1 + k 2 For k = 4, we have: E Ri4 = σ4 22 Γ (3) with Γ (3) = (3 1)! = 2! = 2 So, we have: E Ri4 = σ4 23 = 8σ4 Christophe Hurlin (University of Orléans) Advanced Econometrics - HEC Lausanne November 2013 11 / 68 Solution (cont’d): The variance of Ri2 is equal to: V Ri2 = E Ri4 4σ4 = 8σ4 4σ4 = 4σ4 As a consequence b2 V σ = = = = Christophe Hurlin (University of Orléans) 1 4N 2 1 4N 2 N ∑V Ri2 i =1 N ∑ 4σ4 i =1 N4σ4 4N 2 σ4 N Advanced Econometrics - HEC Lausanne November 2013 12 / 68 Solution (cont’d): To sum up: b2 = σ2 (unbiased estimator) E σ b2 = lim V σ N !∞ σ4 =0 N !∞ N lim b2 is a (weakly) consistent estimator of σ2 : So, the estimator σ p b2 ! σ2 σ Christophe Hurlin (University of Orléans) Advanced Econometrics - HEC Lausanne November 2013 13 / 68 Problem (cont’d) Question 3: Sometimes, the Rayleigh distribution is parametrized by σ (and not by σ2 ) with R Rayleigh (σ) Propose a (weakly) consistent estimator for σ. Christophe Hurlin (University of Orléans) Advanced Econometrics - HEC Lausanne November 2013 14 / 68 Solution: Since σ > 0, a natural estimator for σ is de…ned by: v u p u 1 N t 2 b b2 σ= Ri = σ 2N i∑ =1 b2 is a (weakly) consistent estimator of σ2 : We shown that the estimator σ p b2 ! σ2 σ By applying the Continuous Mapping Theorem (CMP) for the continuous p function g (x ) = x, we get immediately: p or Christophe Hurlin (University of Orléans) b2 ! g σ2 g σ p b!σ σ Advanced Econometrics - HEC Lausanne November 2013 15 / 68 Problem (cont’d) Question 4: For any value of σ2 , what is the …nite sample (or exact b2 de…ned by sampling) distribution of the estimator σ b2 = σ Christophe Hurlin (University of Orléans) 1 2N N ∑ Ri2 i =1 Advanced Econometrics - HEC Lausanne November 2013 16 / 68 Solution: The estimator is de…ned by b2 = σ 1 2N N ∑ Ri2 i =1 We know that R1 , R2 , .., RN are i.i.d. random variables with a Rayleigh distribution. Ri Rayleigh σ2 Reminder: if X and Y are independent and normally distributed N 0, σ2 p random variables, the transformed variable R = X 2 + Y 2 has a Rayleigh distribution. The exact distribution of R 2 = X 2 + Y 2 is unknown (it is not a χ2 ...), b2 is unknown. and as a consequence the …nite sample distribution of σ Christophe Hurlin (University of Orléans) Advanced Econometrics - HEC Lausanne November 2013 17 / 68 Problem (cont’d) Question 5: Write a Matlab code in order to approximate the true b2 for a sample size N = 10, a (unknown) …nite sample distribution of σ true value of σ2 = 16 and by using S = 1, 000 simulations. b2 . (1) Plot an histogram of the 1,000 realisations of the estimator σ (2) plot the Kernel estimator of the density fσb2 (x ) by using the Matlab built-in function ksdensity. Christophe Hurlin (University of Orléans) Advanced Econometrics - HEC Lausanne November 2013 18 / 68 De…nition (Kernel density estimator) Let consider a sample X1 , .., XN , where X has a distribution characterized by the pdf fX (x ) , for x 2 R. A consistent (kernel) estimator of fX (x ) for any x 2 R is given by: 1 b fX ( x ) = λN N ∑K i =1 x xi λ where K (.) denotes a kernel function and λ is bandwidth parameter. p b fX ( x ) ! fX ( x ) 8x 2 R For more details (and a discussion on the optimal choice of λ), see: Lecture notes "Econométrie Non Paramétrique", Hurlin (2008), Master Econométrie and Statistique Appliquée, Université d’Orléans. Christophe Hurlin (University of Orléans) Advanced Econometrics - HEC Lausanne November 2013 19 / 68 De…nition (Kernel function) A kernel function K (u ) satis…es the following properties: (i ) K (u ) 0 R (ii ) K (u ) du = 1 (iii ) K (u ) reaches its maximum for u = 0 and decreases with ju j. (iv ) K (u ) is symmetric, i.e. K (u ) = K ( u ) . Christophe Hurlin (University of Orléans) Advanced Econometrics - HEC Lausanne November 2013 20 / 68 Some examples of Kernel functions u2 2 1 Normal : K (u ) = p exp 2π Triangular : K (u ) = 1 Quartic or BiWeight : K (u ) = Epanechnikov : K (u ) = Triweight : K (u ) = Christophe Hurlin (University of Orléans) u 2 [ 1, 1] 15 1 16 3 1 4 35 1 32 ju j u2 u2 u2 u2R 3 Advanced Econometrics - HEC Lausanne 2 u 2 [ 1, 1] u 2 [ 1, 1] u 2 [ 1, 1] November 2013 21 / 68 Christophe Hurlin (University of Orléans) Advanced Econometrics - HEC Lausanne November 2013 22 / 68 250 200 150 100 50 0 5 Christophe Hurlin (University of Orléans) 10 15 20 25 Advanced Econometrics - HEC Lausanne 30 35 November 2013 23 / 68 0.08 0.07 0.06 0.05 0.04 0.03 0.02 0.01 0 0 5 Christophe Hurlin (University of Orléans) 10 15 20 25 30 Advanced Econometrics - HEC Lausanne 35 40 November 2013 24 / 68 Problem (cont’d) Question 6: For the special case where σ2 = 1, what is the …nite sample b2 de…ned by (or exact sampling) distribution of the estimator σ b2 = σ Christophe Hurlin (University of Orléans) 1 2N N ∑ Ri2 i =1 Advanced Econometrics - HEC Lausanne November 2013 25 / 68 Solution: The estimator is de…ned by b2 = σ 1 2N N ∑ Ri2 i =1 We know that R1 , R2 , .., RN are i.i.d. random variables with Ri Rayleigh (1) p Reminder 1 : if Ri Rayleigh (1), then R = X 2 + Y 2 where X and Y are independent and standard normally distributed N (0, 1) random variables. Reminder 2: if X N (0, 1) , then X 2 χ2 (1) Reminder 3: if X χ2 (v1 ) and Y χ2 (v2 ) , and X and Y are 2 independent, then X + Y χ (v1 + v2 ) Christophe Hurlin (University of Orléans) Advanced Econometrics - HEC Lausanne November 2013 26 / 68 Solution (cont’d): So, if X and Y are independent and standard normally distributed N (0, 1) random variables p Ri = X 2 + Y 2 Rayleigh (1) Ri2 = X 2 + Y 2 χ2 (2) The sum of independent chi-squared distributed random variable has a chi-squared distribution. N ∑ Ri2 χ2 (2N ) i =1 Christophe Hurlin (University of Orléans) Advanced Econometrics - HEC Lausanne November 2013 27 / 68 Solution (cont’d): b2 = 2N σ N ∑ Ri2 χ2 (2N ) i =1 b2 has an In the special case where σ2 = 1, the transformed variable 2N σ exact sampling (…nite sample) distribution that corresponds to a chi-squared distribution with 2N degrees of freedom. Christophe Hurlin (University of Orléans) Advanced Econometrics - HEC Lausanne November 2013 28 / 68 Problem (cont’d) Question 7: Write a Matlab code in order to approximate the true b2 for (unknown) …nite sample distribution of the transformed variable 2N σ a sample size N = 10 in the special case where σ2 = 1 by using S = 10, 000 simulations. (1) Plot the Kernel estimator of the density f2N σb2 (x ) by using the Matlab built-in function ksdensity. (2) Compare this estimated density function to the pdf of a chi-squared distribution with 2N degrees of freedom. Christophe Hurlin (University of Orléans) Advanced Econometrics - HEC Lausanne November 2013 29 / 68 Christophe Hurlin (University of Orléans) Advanced Econometrics - HEC Lausanne November 2013 30 / 68 0.07 Estimated finite sample pdf Theoretical pdf of a chi-squared 0.06 0.05 0.04 0.03 0.02 0.01 0 0 Christophe Hurlin (University of Orléans) 10 20 30 40 50 Advanced Econometrics - HEC Lausanne 60 70 November 2013 31 / 68 Problem (cont’d) b2 ? Question 8: What is the asymptotic distribution of the estimator σ b2 = σ Christophe Hurlin (University of Orléans) 1 2N N ∑ Ri2 i =1 Advanced Econometrics - HEC Lausanne November 2013 32 / 68 Solution: We know that R12 , R22 , .., RN2 are i.i.d. variables with E Ri2 = 2σ2 V Ri2 = 4σ4 Step 1: By applying the Lindberg-Levy univariate Central Limit Theorem (CLT), we get: ! p 1 N 2 d 2 N Ri 2σ ! N 0, 4σ4 N i∑ =1 Christophe Hurlin (University of Orléans) Advanced Econometrics - HEC Lausanne November 2013 33 / 68 Solution (cont’d): Step 2: By de…nition, we have 1 b = σ 2N 2 p N 1 N N ∑ N ∑ Ri2 =g i =1 Ri2 i =1 1 N 2σ 2 ! N ∑ i =1 Ri2 ! d ! N 0, 4σ4 with g (x ) = x /2. So, g (.) is a continuous and continuously di¤erentiable function with g 2σ2 6= 0 and not involving N, then the delta method implies ! ! ! 2 p 1 N 2 ∂g x ( ) d Ri g 2σ2 ! N 0, N g 4σ4 N i∑ ∂x 2 2σ =1 g 2σ2 = 2σ2 = σ2 2 Christophe Hurlin (University of Orléans) ∂g (x ) ∂x = 2σ2 Advanced Econometrics - HEC Lausanne ∂x /2 ∂x = 2σ2 1 2 November 2013 34 / 68 Solution (cont’d): p N g 1 N N ∑ i =1 Ri2 ! g 2σ 2 ! d !N 0, 1 2 2 4σ 4 ! b2 is asymptotically normally distributed The estimator σ p Christophe Hurlin (University of Orléans) b2 N σ d σ2 ! N 0, σ4 Advanced Econometrics - HEC Lausanne November 2013 35 / 68 Problem (cont’d) b2 ? Question 9: What is the asymptotic variance of the estimator σ b2 = σ Christophe Hurlin (University of Orléans) 1 2N N ∑ Ri2 i =1 Advanced Econometrics - HEC Lausanne November 2013 36 / 68 Solution: We know that: p or equivalently b2 N σ b2 σ asy d σ2 ! N 0, σ4 N σ2 , b2 is equal to: The asymptotic variance of σ b2 = Vasy σ Christophe Hurlin (University of Orléans) σ4 N σ4 N Advanced Econometrics - HEC Lausanne November 2013 37 / 68 Problem (cont’d) Question 10: Write a Matlab code in order to approximate the asymptotic distribution of the transformed variable p Z = b2 N σ σ2 σ2 for a sample size N = 10, 000, a true value of σ2 = 16 by using S = 10, 000 simulations. (1) Plot the Kernel estimator of the density fZ (x ) by using the Matlab built-in function ksdensity. (2) Compare this estimated density function to the pdf of a standard normal distribution. Christophe Hurlin (University of Orléans) Advanced Econometrics - HEC Lausanne November 2013 38 / 68 Christophe Hurlin (University of Orléans) Advanced Econometrics - HEC Lausanne November 2013 39 / 68 0.45 0.4 Estimated finite sample pdf Theoretical pdf of a standard normal 0.35 0.3 0.25 0.2 0.15 0.1 0.05 0 -5 Christophe Hurlin (University of Orléans) 0 Advanced Econometrics - HEC Lausanne 5 November 2013 40 / 68 Problem (cont’d) b of Question 11: What is the asymptotic distribution of the estimator σ the parameter σ de…ned by: v u u 1 N b=t σ Ri2 2N i∑ =1 Christophe Hurlin (University of Orléans) Advanced Econometrics - HEC Lausanne November 2013 41 / 68 Solution: Step 1 : We know that: p b > 0, we have: Since σ d b2 N σ v u u 1 b=t σ 2N σ2 ! N 0, σ4 N ∑ Ri2 = i =1 p b2 = g σ b2 σ p where g (x ) = x is a continuous and continuously di¤erentiable function with g σ2 6= 0 and that does not depend on N Christophe Hurlin (University of Orléans) Advanced Econometrics - HEC Lausanne November 2013 42 / 68 Solution (cont’d): Step 2: We have p d b2 N σ σ2 ! N 0, σ4 p The delta method for g (x ) = x implies p with b N g σ g σ 2 2 =σ Christophe Hurlin (University of Orléans) g σ2 ∂g (x ) ∂x d !N σ2 ∂g (x ) ∂x 0, p ∂ x = ∂x σ2 2 σ4 σ2 ! 1 1 = p = 2σ 2 σ2 Advanced Econometrics - HEC Lausanne November 2013 43 / 68 Solution (cont’d): Step 2 (cont’d): p b2 N g σ g σ So, we have: 2 d g σ2 =σ p Christophe Hurlin (University of Orléans) !N ∂g (x ) ∂x b N (σ σ2 d 0, p ∂ x = ∂x σ) ! N ∂g (x ) ∂x 0, σ2 2 σ4 σ2 ! 1 1 = p = 2 2σ 2 σ σ2 4 Advanced Econometrics - HEC Lausanne November 2013 44 / 68 Exercise 2 CAPM Christophe Hurlin (University of Orléans) Advanced Econometrics - HEC Lausanne November 2013 45 / 68 Problem (CAPM) The empirical analogue of the CAPM is given by: = αi + βi rit rft | {z } excess return of security i for time t market excess return for time t where εt is an i.i.d. error term. We assume that e rit = rit E ( εt ) = 0 Christophe Hurlin (University of Orléans) + εt (rmt rft ) | {z } rft e rmt = rmt V ( εt ) = σ 2 rft rmt ) = 0 E ( εt j e Advanced Econometrics - HEC Lausanne November 2013 46 / 68 Problem (CAPM, cont’d) Consider the model e rit = αi + βi e rmt + εt Data: Microsoft, SP500 and Tbill (closing prices) from 11/1/1993 to 04/03/2003 0.10 0.08 RMSFT 0.05 0.04 0.00 0.00 -0.05 -0.04 -0.10 -0.06 -0.04 -0.02 0.00 0.02 0.04 0.06 0.08 -0.08 500 RSP500 Christophe Hurlin (University of Orléans) 1000 RSP500 Advanced Econometrics - HEC Lausanne 1500 2000 RMSFT November 2013 47 / 68 Problem (CAPM, cont’d) We consider the CAPM model rewritten as follows e rit = xt> β + εt t = 1, ..T where xt = (1 e rmt )> is 2 1 vector of random variables, β = (αi βi )> is 2 1 vector of parameters, and where the error term εt satis…es rmt ) = 0. E (εt ) = 0, V (εt ) = σ2 and E ( εt j e Christophe Hurlin (University of Orléans) Advanced Econometrics - HEC Lausanne November 2013 48 / 68 Problem (CAPM, cont’d) Question 1: show that the OLS estimator ! 1 T b = ∑ xt xt> β t =1 satis…es p b T β Christophe Hurlin (University of Orléans) d T ∑ xt erit t =1 β0 ! N 0, σ2 E 1 Advanced Econometrics - HEC Lausanne ! xt> xt November 2013 49 / 68 Solution: 1 let us rewrite the OLS estimator as: ! 1 ! T T > b = ∑ xt x β ∑ xt erit = β0 + t t =1 2 b T β Christophe Hurlin (University of Orléans) ∑ t =1 b Normalize the vector β p T β0 = xt xt> t =1 ! 1 T ∑ xt εt t =1 ! β0 1 T T ∑ xt xt> t =1 ! 1 Advanced Econometrics - HEC Lausanne p 1 T T T ∑ xt εt t =1 ! November 2013 50 / 68 Solution (cont’d): 3. Using the WLLN and the CMP: 1 T T ∑ xt xt> t =1 ! 1 p ! E 1 xt xt> 4. Using the CLT: p T 1 T T ∑ xt εt t =1 E (xt εt ) ! d ! N (0, V (xt εt )) with E ( εt j e rmt ) = 0 and E ( εt j 1) = E (εt ) = 0 =) E (xt εt ) = 0 and V (xt εt ) = E xt εt εt xt> = E E xt εt εt xt> xt = E xt V ( εt j xt ) xt> = σ2 E xt xt> Christophe Hurlin (University of Orléans) Advanced Econometrics - HEC Lausanne November 2013 51 / 68 Solution (cont’d): So, we have 1 T p T 1 T Christophe Hurlin (University of Orléans) T ∑ xt xt> t =1 T ∑ xt εt t =1 ! ! 1 p ! E 1 xt xt> d ! N 0, σ2 E xt xt> Advanced Econometrics - HEC Lausanne November 2013 52 / 68 Solution (cont’d): By using the Slutsky’s theorem (for a convergence in distribution), we have: ! 1 ! T p 1 T p 1 d > b β = T β xt xt T ∑ xt µt ! N (Π, Ω) 0 T t∑ T =1 t =1 with Π=E Ω=E 1 xt xt> 1 xt xt> σ2 E xt xt> 0=0 E 1 xt xt> = σ2 E 1 xt xt> Finally, we have: p b T β Christophe Hurlin (University of Orléans) d β0 ! N 0, σ2 E 1 xt xt> Advanced Econometrics - HEC Lausanne November 2013 53 / 68 Problem (CAPM, cont’d) Question 2: What is the asymptotic variance-covariance matrix of the b? OLS estimator β b= β Christophe Hurlin (University of Orléans) T ∑ t =1 xt xt> ! 1 T ∑ xt erit t =1 Advanced Econometrics - HEC Lausanne ! November 2013 54 / 68 Solution: We shown that p or equivalently b T β b β d β0 ! N 0, σ2 E asy N β0 , σ2 E T 1 1 xt xt> xt xt> b is equal to: The asymptotic variance-covariance matrix of β 2 b = σ E Vasy β T| Christophe Hurlin (University of Orléans) 1 xt xt> {z } 2 2 Advanced Econometrics - HEC Lausanne November 2013 55 / 68 Remarks: 1 2 The asymptotic variance covariance matrix is a 2 matrix: 0 2 Vasy (b α) σ b = E 1 xt xt> = @ Vasy β T β, b α cov b Since xt = (1 e rmt )> , we have: E xt xt> = E Christophe Hurlin (University of Orléans) 1 e rmt 2 e rmt e rmt = 2 symmetric α, b β cov b β Vasy b 1 A 1 E (e rmt ) 2 E (e rmt ) E e rmt Advanced Econometrics - HEC Lausanne November 2013 56 / 68 Problem (CAPM, cont’d) Question 3: Let us consider a consistent estimator of σ2 de…ned by: b2 = σ 1 T T ∑ bε2t = 2 t =1 1 T T 2 ∑ t =1 e rit b xt> β 2 b ? Propose a consistent estimator of the asymptotic variance Vasy β Christophe Hurlin (University of Orléans) Advanced Econometrics - HEC Lausanne November 2013 57 / 68 Solution: We know that 2 b = σ E Vasy β T Using the LLN, if xt is i.i.d., we get: 1 T T p xt xt> 1 ∑ xt xt> ! E xt xt> ! 1 t =1 Using the CMP: 1 T Christophe Hurlin (University of Orléans) T ∑ t =1 xt xt> 1 p !E xt xt> Advanced Econometrics - HEC Lausanne November 2013 58 / 68 Solution (cont’d): So, we have 2 b = σ E Vasy β T and 1 T T ∑ xt xt> t =1 ! 1 xt xt> 1 p !E 1 xt xt> p By using the Slutsky’s theorem: b2 σ T 1 T T ∑ xt xt> t =1 Christophe Hurlin (University of Orléans) ! b2 ! σ2 σ 1 p ! σ2 E T 1 b xt xt> = Vasy β Advanced Econometrics - HEC Lausanne November 2013 59 / 68 Solution (cont’d): b is de…ned by A consistent estimator of the asymptotic variance Vasy β b asy V Or equivalently by b2 b = σ β T 1 T b =σ b asy β b2 V with b2 = σ 1 T Christophe Hurlin (University of Orléans) T ∑ bε2t = 2 t =1 1 T T ∑ xt xt> t =1 T ∑ xt xt> t =1 ! T 2 ∑ t =1 ! e rit Advanced Econometrics - HEC Lausanne 1 1 b xt> β 2 November 2013 60 / 68 Remark: Since xt = (1 e rmt )> : b =σ b asy β b2 V T ∑ xt xt> = t =1 Christophe Hurlin (University of Orléans) T ∑ t =1 T rmt ∑Tt=1 e xt xt> ! 1 rmt ∑Tt=1 e 2 rmt ∑Tt=1 e Advanced Econometrics - HEC Lausanne November 2013 61 / 68 Problem (CAPM, cont’d) Question 4: Using the excel …le capm.xls, write a Matlab code to estimate the beta and the alpha for MSFT. Compare your results with the following table of estimation results (Eviews). Christophe Hurlin (University of Orléans) Advanced Econometrics - HEC Lausanne November 2013 62 / 68 Christophe Hurlin (University of Orléans) Advanced Econometrics - HEC Lausanne November 2013 63 / 68 Perfect.... Christophe Hurlin (University of Orléans) Advanced Econometrics - HEC Lausanne November 2013 64 / 68 Problem (CAPM, cont’d) Question 5: Using the excel …le capm.xls, write a Matlab code (1) to estimate the variance of the error term εt b (2) to estimate the asymptotic standard errors of the estimators β (3) Compare your results with the table of estimation results (Eviews). Christophe Hurlin (University of Orléans) Advanced Econometrics - HEC Lausanne November 2013 65 / 68 Christophe Hurlin (University of Orléans) Advanced Econometrics - HEC Lausanne November 2013 66 / 68 Perfect too.... Christophe Hurlin (University of Orléans) Advanced Econometrics - HEC Lausanne November 2013 67 / 68 End of Exercices - Chapter 1 Christophe Hurlin (University of Orléans) Christophe Hurlin (University of Orléans) Advanced Econometrics - HEC Lausanne November 2013 68 / 68