Testing Downside-Risk Efficiency Under Distress

Size: px
Start display at page:

Download "Testing Downside-Risk Efficiency Under Distress"

Transcription

1 Testing Downside-Risk Efficiency Under Distress Jesus Gonzalo Universidad Carlos III de Madrid Jose Olmo City University of London XXXIII Simposio Analisis Economico 1

2 Some key lines Risk vs Uncertainty. LP M vs Variance. LP M q = R p <τ P p (τ R p ) q. Portfolio optimization: min {w} s.t. LP M q (τ, w) = m w j R j = Rp; j=1 R p <τ P p (τ R p ) q m w j = 1, w j > 0. j=1 Utility function compatible with downside risk approach: U(R p ; q, τ) = R p k(τ R p ) q I(R p τ). XXXIII Simposio Analisis Economico 2

3 How do we compare portfolios? By Stochastic Dominance (SD). There is a close connection between LP M q measures and q stochastic dominance. Need for testing q-stochastic dominance. Thanks to a decomposition that we propose for LP M q we develop a new test for SD with a simpler asymptotic theory. When all of this is crucial? In distress periods: (R 1 u, R 2 u,..., R m u). We re-do the theory on stochastic dominance and hypothesis testing for distress periods. XXXIII Simposio Analisis Economico 3

4 Motivation Mean-Variance investment strategies are efficient only under restrictive assumptions. Mean-Variance investment strategies are not necessarily efficient under distress episodes of the market. Mean-Risk efficient sets are alternative investment strategies for investors concerned with downside risk, and where this is defined by a threshold. Downside-risk efficient strategies are therefore dependent on the level of investor s risk aversion and can vary heavily with the choice of the threshold. Stochastic dominant portfolios, instead, are optimal for every point in the domain of the relevant distributions. The concept of stochastic dominance can be extended to subsets of the domain of the distribution of the data. In our case to the left tail of the distribution (distress regimes). XXXIII Simposio Analisis Economico 4

5 Contributions Main aim of a diversified investor: To derive investment strategies that minimize risk for target return levels, and, ideally, under comovement episodes of the market. The benchmark being Mean-Variance strategies. In order to do this we need: To decompose the downside risk in a portfolio in downside risk due to the presence of comovements between the assets and due to the choice of optimal weights. To determine the relationship between mean-risk and stochastic dominance efficient sets under distress episodes of the market. To develop a novel test statistic for testing stochastic dominance that makes allowance for testing orders of dominance higher than zero and for general forms of dependence between portfolios. XXXIII Simposio Analisis Economico 5

6 Outline Relationship between mean-risk and stochastic dominance efficiency, and its extension to market distress. Develop a hypothesis test for stochastic dominance and conditional stochastic dominance. A Monte-Carlo experiment to study finite-sample performance of the test. An empirical comparison between mean-variance and stochastic dominance investment strategies. Conclusions. XXXIII Simposio Analisis Economico 6

7 Mean-Risk and Stochastic Dominance Efficient Sets Risk is based on the idea of dread events, see Hogan and Warren (1974), Bawa (1975), Arzac and Bawa (1977), or Bawa and Lindenberg (1977). The utility function corresponding to these investors is U(R p ; q, τ) = R p k(τ R p ) q I(R p τ), (1) with R p return on a portfolio P ; τ target return; k scaling parameter, I( ) an indicator function, and q investor s risk aversion level. The relevant risk measure is LP M q (τ) = τ (τ R p ) q df (x), (2) with F (x) = P (R p x). Result 1 (Fishburn, 1977): Portfolio A dominates Portfolio B in the mean-risk model if and only if µ(a) µ(b) and LP M q (A) LP M q (B) for q 0, with at least one strict inequality. XXXIII Simposio Analisis Economico 7

8 Stochastic Dominance Result 2 (Fishburn, 1977): A first stochastic dominates (FSD) B if and only if F A F B LP M0 B (τ) for all τ R, and LP M A 0 (τ) A second stochastic dominates (SSD) B if and only if F A F B and LP M A 1 (τ) LP M B 1 (τ) for all τ R, A third stochastic dominates (TSD) B if and only if F A F B, µ(a) µ(b), and LP M A 2 (τ) LP M B 2 (τ) for all τ R, with F A and F B the distribution functions of two portfolios A and B. Moreover, this author also shows that stochastic dominance of order one is sufficient for mean-risk efficiency for q 0 (risk-neutral and risk-averse investors). Also, stochastic dominance of order two is sufficient for mean-risk efficiency for q 1 (risk-averse investors). We extend these results to Distress periods. XXXIII Simposio Analisis Economico 8

9 Definition of a Distress Measure In our framework we will identify the presence of distress comovements with the event: {R 1 u,..., R m u} with probability P {R 1 u,..., R m u} := λ(u). The relevant new risk measure is LP M P q,u(τ) = τ (τ x) q df u (x), (3) where F u (x) := P {R p x R 1 u,..., R m u}. proposition 1: Let LP M P q ( ) and LP M P q,u( ) for q 0 be the downside risk measures defined in (2) and (3) respectively. Then, under some regularity assumptions, LP M P q (τ) = E[(τ R p ) q R p τ]lp M P 0 (τ), (4) and LP M P q,u(τ) = E[(τ R p ) q R p τ, R 1 u,..., R m u]lp M P 0,u(τ). (5) XXXIII Simposio Analisis Economico 9

10 Corollary 1: The unconditional downside risk measure of interest can be decomposed in terms of the downside risk measures conditional on distress comovements and noncomovements regimes. LP M q (τ) = λ(u)γ q,u (τ)lp M q,u (τ) + (1 λ(u))γ q,ū (τ)lp M q,ū (τ), (6) with γ q,u (τ) = E[(τ R p ) q R p τ] E[(τ R p ) q R p τ,r 1 u,r 2 u,...,r m u], γ q,ū (τ) = E[(τ R p ) q R p τ] E[(τ R p ) q R p τ,r 1 >u or R 2 >u or...or R m >u], and LP M q,ū the risk measure conditional on {R 1 u, R 2 u,..., R m u} C. Remark: This corollary shows that A more efficient than B unconditionally does not imply that A more efficient than B under distress periods. In fact, there can other allocations more efficient under distress episodes of the market. XXXIII Simposio Analisis Economico 10

11 Conditional Stochastic Dominance Under Distress With this decomposition in place and the conditional downside-risk measure we can define conditional stochastic dominance. Definition 1: A first conditional stochastic dominates (FCSD) B if and only if F A u F B u and LP M A 0,u(τ) LP M B 0,u(τ) for all τ u, A second conditional stochastic dominates (SCSD) B if and only if F A u F B u and LP M A 1,u(τ) LP M B 1,u(τ) for all τ u, A third conditional stochastic dominates (TCSD) B if and only if Fu A µ u (B), and LP M2,u(τ) A LP M2,u(τ) B for all τ u, F B u, µ u (A) XXXIII Simposio Analisis Economico 11

12 theorem 1: If A FCSD B then A dominates B in the mean-risk model defined by LP M q,u measures for all q 0. If A SCSD B then A dominates B in the mean-risk model defined by LP M q,u measures for all q 1, except when µ u (A) = µ u (B) and LP M A q,u(τ) = LP M B q,u(τ) for all τ u. If A TCSD B then A dominates B in the mean-risk model defined by LP M q,u measures for all q 2, except when µ u (A) = µ u (B) and LP M A q,u(τ) = LP M B q,u(τ) for all τ u. In order to make the conditions stated before statistically testable we develop in what follows hypothesis tests for stochastic and conditional stochastic dominance of different orders. XXXIII Simposio Analisis Economico 12

13 Hypothesis Tests for Stochastic Dominance Our test statistic is of Kolmogorov-Smirnov type and shares the spirit of the test statistics proposed in in McFadden (1989), Anderson (1996), Davidson and Duclos (2000), Barret and Donald (2003) or Linton, Maasoumi and Whang (2005) among others. The results in Fishburn (1977) allow us to focus on the hypothesis test { H0,γ : LP Mγ A (τ) LP Mγ B (τ), for all τ R, H 1,γ : LP Mγ A (τ) > LP Mγ B (τ), for some τ R, (7) rather than on the strict inequality, for testing first (γ = 0), second (γ = 1) and third (γ = 2) stochastic dominance between two portfolios A and B. Alternatively, one can write the test as { H0,γ : D γ (τ) 0, for all τ R, H 1,γ : D γ (τ) > 0, for some τ R, (8) with D γ (τ) := LP M A γ (τ) LP M B γ (τ). XXXIII Simposio Analisis Economico 13

14 Asymptotic theory theorem 2: Under some regularity assumptions, n sup ( D γ (τ) D γ (τ)) τ R d sup G γ (τ), (9) τ R with G γ (τ) a Gaussian process with zero mean and covariance function given by E[G γ (τ i )G γ (τ j )] = ( k2γ(τ A i τ j )F A (τ i τ j ) kγ A (τ i )F A (τ i )kγ A (τ j )F A (τ j ) ) ( + k B 2γ (τ i τ j )F B (τ i τ j ) kγ B (τ i )F B (τ i )kγ B (τ j )F B (τ j ) ) ( k A,B γ (τ i, τ j )F A,B (τ i, τ j ) kγ A (τ i )F A (τ i )kγ B (τ j )F B (τ j ) ) ( k A,B γ (τ j, τ i )F A,B (τ j, τ i ) k A γ (τ j )F A (τ j )k B γ (τ i )F B (τ i ) ), (10) for all τ i, τ j R. Notation: F A,B (τ, τ) := P (R A p τ, RB p τ), ki γ (τ) = E[(τ Ri p )γ R i p and k A,B γ (τ, τ) = E[(τ R A p )γ (τ R B p )γ R A p τ, RB p τ] τ] with i = A, B XXXIII Simposio Analisis Economico 14

15 A new test statistic for H 0,γ Our family of test statistics is defined by T n,γ := n sup τ R D γ (τ). The null hypothesis is the equality of functions LP M A γ (τ) = LP M B γ (τ) for every τ R. Under some regularity assumptions, and H 0,γ, T n,γ d sup G γ (τ). (11) τ R Further, the asymptotic critical values of these tests indexed by γ are given by c γ (1 α) := inf x R {x P with α denoting the significance level. ( ) sup G γ (τ) x τ R 1 α}, (12) XXXIII Simposio Analisis Economico 15

16 Consistency of the tests proposition 3: Under some regularity assumptions and the test statistic T n,γ, then: (i) Under H 0,γ, lim P (reject H 0,γ) = lim P (T n,γ > c γ (1 α)) α, (13) n n with equality when F A (τ) = F B (τ) for every τ R. (ii) If H 0,γ is false, lim P (reject H 0,γ) = lim P (T n,γ > c γ (1 α)) = 1. (14) n n XXXIII Simposio Analisis Economico 16

17 P-value approximations for stochastic dominance tests of orders higher than zero 2,..., x(j) n ), j = 1,..., be a collection of random samples of dimension n 2 drawn from a bivariate distribution F A,B (τ, τ). Let T n,γ (j) be the test statistic associated to each sample, and c (j) γ (1 α) the corresponding critical proposition 4: Let x (j) n := (x (j) 1, x(j) values obtained from the corresponding estimated functional of G γ. Then (i) Under H 0,γ, lim P (reject H (j) 0,γ) = lim P (T n,γ > c (1) γ (1 α)) α, (15) n n almost surely for every random sample x (j) n, and with equality when F A (τ) = F B (τ). (ii) If H 0,γ is false, lim P (reject H (j) 0,γ) = lim P (T n,γ > c (1) γ (1 α)) = 1, n n almost surely for every random sample x (j) n. XXXIII Simposio Analisis Economico 17

18 Hypothesis Test for Stochastic Dominance Under Distress { H0,γ,u : D γ,u (τ) 0, for all τ R, (16) H 1,γ,u : D γ,u (τ) > 0, for some τ R, where D γ,u (τ) = LP M A γ,u(τ) LP M B γ,u(τ). theorem 3: Under some regularity assumptions, nu sup ( D γ,u (τ) D γ,u (τ)) τ (,u] d sup G γ,u (τ), (17) τ (,u] with G γ,u (τ) a Gaussian process with zero mean and covariance function given by E[G γ,u (τ i )G γ,u (τ j )] = ( k2γ,u(τ A i τ j )Fu A (τ i τ j ) kγ,u(τ A i )Fu A (τ i )kγ,u(τ A j )Fu A (τ j ) ) ( + k B 2γ,u (τ i τ j )Fu B (τ i τ j ) kγ,u(τ B i )Fu B (τ i )kγ,u(τ B j )Fu B (τ j ) ( k A,B γ,u (τ i, τ j )Fu A,B (τ i, τ j ) kγ,u(τ A i )Fu A (τ i )kγ,u(τ B j )Fu B (τ j ) ) ( k A,B γ,u (τ j, τ i )Fu A,B (τ j, τ i ) kγ,u(τ A j )Fu A (τ j )kγ,u(τ B i )Fu B (τ i ) ), for all τ i, τ j u. XXXIII Simposio Analisis Economico 18

19 The family of test statistics for H 0,γ,u are T nu,γ := n u sup τ (,u] D γ,u (τ). Under some regularity assumptions, and H 0,γ,u, with u R, satisfy T nu,γ d sup G γ,u (τ). τ (,u] The asymptotic critical values of these tests are given by c γ,u (1 α) := inf x R {x P ( sup G γ,u (τ) x τ (,u] ) 1 α}, (18) with α denoting the significance level. Simulation procedures as a p-value transformation or bootstrap can be proposed to approximate the critical value of the test. Alternatively, we propose to estimate the asymptotic covariance function from the data and approximate the relevant critical value by Monte-Carlo simulation of the restricted supremum of the estimated gaussian process. XXXIII Simposio Analisis Economico 19

20 Finite Sample Performance: Size In Table 1 empirical size of the different hypothesis tests H 0,γ, γ = 0, 1: First and Second Order Stochastic Dominance. In Table 2 empirical size of the different hypothesis tests H 0,γ,u, γ = 0, 1, u = 0: First and Second Order Stochastic Dominance Under Distress. Two portfolios A and B are generated from a Student-t distribution with ν degrees of freedom. Different scenarios of dependence between portfolios are analyzed by increasing the correlation ρ between marginal Student-t distributions. The accuracy of the approximation given by the estimated critical values is assessed by using different sample sizes: n = 50, 100, 500. For the conditional test under distress we use n = 200, 400, 2000 in order to have roughly the same sample sizes as for the unconditional experiment: n u = 50, 100, 500. XXXIII Simposio Analisis Economico 20

21 Table 1. Empirical size for H 0,γ, γ = 0, 1 for a standardized bivariate Student-t with ν = 5 d.f and correlation ρ. Gp : asymptotic p-value, p : Multiplier method p-value. n sample size. B = 1000 Monte-Carlo simulations to approximate asymptotic critical value. mc = 500 Monte-Carlo iterations to approximate the nominal size. m = 100 partitions of the real line to generate observations from the corresponding asymptotic Gaussian process with a given covariance matrix. ν = 5 Method γ = 0 γ = 1 ρ = 0 10% 5% 1% 10% 5% 1% n = 50 Gp-value p-value n = 100 Gp-value p-value n = 500 Gp-value p-value XXXIII Simposio Analisis Economico 21

22 ν = 5 Method γ = 0 γ = 1 ρ = % 5% 1% 10% 5% 1% n = 50 Gp-value p-value n = 100 Gp-value p-value n = 500 Gp-value p-value ν = 5 Method γ = 0 γ = 1 ρ = % 5% 1% 10% 5% 1% n = 50 Gp-value p-value n = 100 Gp-value p-value n = 500 Gp-value p-value XXXIII Simposio Analisis Economico 22

23 Size Conditional Test Under Distress Table 2. Same as before. Threshold value: u = 0. ν = 5 Method γ = 0 γ = 1 ρ = 0 10% 5% 1% 10% 5% 1% n = 200 Gp-value (n u 50) p-value n = 400 Gp-value (n u 100) p-value n = 2000 Gp-value (n u 500) p-value ν = 5 Method γ = 0 γ = 1 ρ = % 5% 1% 10% 5% 1% n = 200 Gp-value (n u 50) p-value n = 400 Gp-value (n u 100) p-value n = 2000 Gp-value (n u 500) p-value XXXIII Simposio Analisis Economico 23

24 ν = 5 Method γ = 0 γ = 1 ρ = % 5% 1% 10% 5% 1% n = 200 Gp-value (n u 50) p-value n = 400 Gp-value (n u 100) p-value n = 2000 Gp-value (n u 500) p-value XXXIII Simposio Analisis Economico 24

25 Finite-Sample Performance: Power In Table 3 empirical power of the different hypothesis tests H 0,γ, γ = 0, 1: First and Second Order Stochastic Dominance. In Table 4 empirical power of the different hypothesis tests H 0,γ,u, γ = 0, 1, u = 0: First and Second Order Stochastic Dominance Under Distress. We generate two portfolios A and B, where F B freedom, and F A (τ) = F B (τ) + cf B (τ) n alternatives. is a Student-t with ν degrees of with c = 0.5, 1, 5, characterizing a family of local Different scenarios of dependence between portfolios are analyzed by increasing the correlation ρ between marginal Student-t distributions. The accuracy of the approximation given by the estimated critical values is assessed by using different sample sizes: n = 50, 100, 500. XXXIII Simposio Analisis Economico 25

26 Table 3. Empirical power for H 0,γ, γ = 0, 1. Alternative hypotheses given by F A (τ) = F B (τ)+ cfb (τ) n with F B and f B a Student-t distribution and density function with ν = 5 and c = 0.5, 1, 5. The correlation parameter is rho. n sample size. B = 1000 Monte-Carlo simulations to approximate asymptotic critical value. mc = 500 Monte-Carlo iterations to approximate the nominal size. m = 100 partitions of the real line to generate observations from the corresponding asymptotic Gaussian process with a given covariance matrix. ν = 5, α = 0.05 γ = 0 γ = 1 ρ = 0 / c = n = n = n = XXXIII Simposio Analisis Economico 26

27 ν = 5, α = 0.05 γ = 0 γ = 1 ρ = 0.4 / c = n = n = n = ρ = 0.8 / c = n = n = n = XXXIII Simposio Analisis Economico 27

28 Power Conditional Test Under Distress Table 4. Empirical power for H 0,γ,u, γ = 0, 1, u = 0. Alternative hypotheses given by F A (τ) = F B (τ)+ cfb (τ) n with F B and f B a Student-t distribution and density function with ν = 5 and c = 0.5, 1, 5. The correlation parameter is rho. n sample size. B = 1000 Monte-Carlo simulations to approximate asymptotic critical value. mc = 500 Monte-Carlo iterations to approximate the nominal size. m = 100 partitions of the real line to generate observations from the corresponding asymptotic Gaussian process with a given covariance matrix. ν = 5, α = 0.05 γ = 0 γ = 1 ρ = 0 / c = n = n = n = XXXIII Simposio Analisis Economico 28

29 ν = 5, α = 0.05 γ = 0 γ = 1 ρ = 0.4 / c = n = n = n = ρ = 0.8 / c = n = n = n = XXXIII Simposio Analisis Economico 29

30 An Empirical Study of Mean-risk Efficiency Consider a portfolio of risky and heavily traded stocks in the US economy given by Microsoft (MSFT), General Electric (GE), Bank of America Corporation (BAC) and Verizon Communications (VZ) spanning the period 02/01/ /12/ u= LPM 0 (τ) 0.5 LPM 0,u (τ) τ τ Left chart: No possible ranking of portfolios in terms of first order stochastic dominance. Right chart (Under distress): This plot tells a different story... XXXIII Simposio Analisis Economico 30

31 This is formalized with the results below of the tests corresponding to the three possible combinations between portfolios. Unconditional test for First stochastic dominance: H 0,0 : A B A : LP M0 P LP M0 P - σp 2 1 A : σp Unconditional test for Second Stochastic Dominance: H 0,1 : A B A : LP M0 P LP M0 P - σp 2 1 A : σp Conditional test for First Stochastic Dominance Under Distress: H 0,0,0 : A B A : LP M0,0 P LP M0,0 P - σp 2 0 A : σp XXXIII Simposio Analisis Economico 31

32 Conclusions In this paper we have extended the concept of stochastic dominance to define it under distress, and have exploited the relationship between mean-risk efficiency and stochastic dominance to derive efficient portfolios in distress episodes of the market. The efficiency of these portfolios depends on several factors; namely, the order of conditional stochastic dominance, the degree of risk aversion of the investor and the definition of the threshold defining distress comovements. The asymptotic distribution of our tests for stochastic dominance conditional on distress takes a simple and estimable form for any order of dominance, and more importantly, unlike most of the seminal papers in the literature, our method makes allowance for general forms of dependence between portfolios. XXXIII Simposio Analisis Economico 32

33 Our method to approximate the asymptotic critical value confirm our asymptotic theory and show the limitations of the p-value transformation under dependence between portfolios to derive the correct asymptotic critical value. This phenomenon is magnified for stochastic dominance under Distress periods. Finally, the empirical study makes us conclude that investment strategies alternative to mean-variance methods need to be carefully considered when financial markets go through distress episodes. Further, the adequacy of different investment strategies should be assessed by means of statistical tests for stochastic dominance. Future research: Tests of contagion between markets based on tail probabilities. Dynamic tests based on realized lower partial moments. Endogenous determination of threshold u defining distress commovements. XXXIII Simposio Analisis Economico 33

Downside Risk Efficiency Under Market Distress

Downside Risk Efficiency Under Market Distress Working Paper 09-44 Departamento de Economía Economic Series 23) Universidad Carlos III de Madrid June 2009 Calle Madrid, 126 28903 Getafe Spain) Fax 34) 916249875 Downside Risk Efficiency Under Market

More information

CONTAGION VERSUS FLIGHT TO QUALITY IN FINANCIAL MARKETS

CONTAGION VERSUS FLIGHT TO QUALITY IN FINANCIAL MARKETS EVA IV, CONTAGION VERSUS FLIGHT TO QUALITY IN FINANCIAL MARKETS Jose Olmo Department of Economics City University, London (joint work with Jesús Gonzalo, Universidad Carlos III de Madrid) 4th Conference

More information

Robust Backtesting Tests for Value-at-Risk Models

Robust Backtesting Tests for Value-at-Risk Models Robust Backtesting Tests for Value-at-Risk Models Jose Olmo City University London (joint work with Juan Carlos Escanciano, Indiana University) Far East and South Asia Meeting of the Econometric Society

More information

Gaussian Slug Simple Nonlinearity Enhancement to the 1-Factor and Gaussian Copula Models in Finance, with Parametric Estimation and Goodness-of-Fit

Gaussian Slug Simple Nonlinearity Enhancement to the 1-Factor and Gaussian Copula Models in Finance, with Parametric Estimation and Goodness-of-Fit Gaussian Slug Simple Nonlinearity Enhancement to the 1-Factor and Gaussian Copula Models in Finance, with Parametric Estimation and Goodness-of-Fit Tests on US and Thai Equity Data 22 nd Australasian Finance

More information

Dependence. Practitioner Course: Portfolio Optimization. John Dodson. September 10, Dependence. John Dodson. Outline.

Dependence. Practitioner Course: Portfolio Optimization. John Dodson. September 10, Dependence. John Dodson. Outline. Practitioner Course: Portfolio Optimization September 10, 2008 Before we define dependence, it is useful to define Random variables X and Y are independent iff For all x, y. In particular, F (X,Y ) (x,

More information

Income-Related Health Transfers Principles and Orderings of Joint Distributions of Income and Health

Income-Related Health Transfers Principles and Orderings of Joint Distributions of Income and Health Income-Related Health Transfers Principles and Orderings of Joint Distributions of Income and Health Mohamed Khaled University of Queensland Paul Makdissi University of Ottawa November 216 Myra Yazbeck

More information

Robustness and bootstrap techniques in portfolio efficiency tests

Robustness and bootstrap techniques in portfolio efficiency tests Robustness and bootstrap techniques in portfolio efficiency tests Dept. of Probability and Mathematical Statistics, Charles University, Prague, Czech Republic July 8, 2013 Motivation Portfolio selection

More information

Dependence. MFM Practitioner Module: Risk & Asset Allocation. John Dodson. September 11, Dependence. John Dodson. Outline.

Dependence. MFM Practitioner Module: Risk & Asset Allocation. John Dodson. September 11, Dependence. John Dodson. Outline. MFM Practitioner Module: Risk & Asset Allocation September 11, 2013 Before we define dependence, it is useful to define Random variables X and Y are independent iff For all x, y. In particular, F (X,Y

More information

Econometrica, Vol. 71, No. 1 (January, 2003), CONSISTENT TESTS FOR STOCHASTIC DOMINANCE. By Garry F. Barrett and Stephen G.

Econometrica, Vol. 71, No. 1 (January, 2003), CONSISTENT TESTS FOR STOCHASTIC DOMINANCE. By Garry F. Barrett and Stephen G. Econometrica, Vol. 71, No. 1 January, 2003), 71 104 CONSISTENT TESTS FOR STOCHASTIC DOMINANCE By Garry F. Barrett and Stephen G. Donald 1 Methods are proposed for testing stochastic dominance of any pre-specified

More information

Financial Econometrics and Quantitative Risk Managenent Return Properties

Financial Econometrics and Quantitative Risk Managenent Return Properties Financial Econometrics and Quantitative Risk Managenent Return Properties Eric Zivot Updated: April 1, 2013 Lecture Outline Course introduction Return definitions Empirical properties of returns Reading

More information

If we want to analyze experimental or simulated data we might encounter the following tasks:

If we want to analyze experimental or simulated data we might encounter the following tasks: Chapter 1 Introduction If we want to analyze experimental or simulated data we might encounter the following tasks: Characterization of the source of the signal and diagnosis Studying dependencies Prediction

More information

Testing for Stochastic Dominance Efficiency

Testing for Stochastic Dominance Efficiency Testing for Stochastic Dominance Efficiency Olivier SCAILLET HEC, Universty of Geneva and FAME Nikolas TOPALOGLOU HEC, University of Geneva Research Paper N 154 July 2005 FAME - International Center for

More information

Comparing downside risk measures for heavy tailed distributions

Comparing downside risk measures for heavy tailed distributions Comparing downside risk measures for heavy tailed distributions Jon Danielsson Bjorn N. Jorgensen Mandira Sarma Casper G. de Vries March 6, 2005 Abstract In this paper we study some prominent downside

More information

When is a copula constant? A test for changing relationships

When is a copula constant? A test for changing relationships When is a copula constant? A test for changing relationships Fabio Busetti and Andrew Harvey Bank of Italy and University of Cambridge November 2007 usetti and Harvey (Bank of Italy and University of Cambridge)

More information

Testing for Bivariate Stochastic Dominance. Using Inequality Restrictions

Testing for Bivariate Stochastic Dominance. Using Inequality Restrictions Testing for Bivariate Stochastic Dominance Using Inequality Restrictions Thanasis Stengos and Brennan S. Thompson May 31, 2011 Abstract In this paper, we propose of a test of bivariate stochastic dominance

More information

Miloš Kopa. Decision problems with stochastic dominance constraints

Miloš Kopa. Decision problems with stochastic dominance constraints Decision problems with stochastic dominance constraints Motivation Portfolio selection model Mean risk models max λ Λ m(λ r) νr(λ r) or min λ Λ r(λ r) s.t. m(λ r) µ r is a random vector of assets returns

More information

A contamination model for approximate stochastic order

A contamination model for approximate stochastic order A contamination model for approximate stochastic order Eustasio del Barrio Universidad de Valladolid. IMUVA. 3rd Workshop on Analysis, Geometry and Probability - Universität Ulm 28th September - 2dn October

More information

EC2001 Econometrics 1 Dr. Jose Olmo Room D309

EC2001 Econometrics 1 Dr. Jose Olmo Room D309 EC2001 Econometrics 1 Dr. Jose Olmo Room D309 J.Olmo@City.ac.uk 1 Revision of Statistical Inference 1.1 Sample, observations, population A sample is a number of observations drawn from a population. Population:

More information

Markowitz Efficient Portfolio Frontier as Least-Norm Analytic Solution to Underdetermined Equations

Markowitz Efficient Portfolio Frontier as Least-Norm Analytic Solution to Underdetermined Equations Markowitz Efficient Portfolio Frontier as Least-Norm Analytic Solution to Underdetermined Equations Sahand Rabbani Introduction Modern portfolio theory deals in part with the efficient allocation of investments

More information

H 2 : otherwise. that is simply the proportion of the sample points below level x. For any fixed point x the law of large numbers gives that

H 2 : otherwise. that is simply the proportion of the sample points below level x. For any fixed point x the law of large numbers gives that Lecture 28 28.1 Kolmogorov-Smirnov test. Suppose that we have an i.i.d. sample X 1,..., X n with some unknown distribution and we would like to test the hypothesis that is equal to a particular distribution

More information

A simple nonparametric test for structural change in joint tail probabilities SFB 823. Discussion Paper. Walter Krämer, Maarten van Kampen

A simple nonparametric test for structural change in joint tail probabilities SFB 823. Discussion Paper. Walter Krämer, Maarten van Kampen SFB 823 A simple nonparametric test for structural change in joint tail probabilities Discussion Paper Walter Krämer, Maarten van Kampen Nr. 4/2009 A simple nonparametric test for structural change in

More information

Introduction to Algorithmic Trading Strategies Lecture 10

Introduction to Algorithmic Trading Strategies Lecture 10 Introduction to Algorithmic Trading Strategies Lecture 10 Risk Management Haksun Li haksun.li@numericalmethod.com www.numericalmethod.com Outline Value at Risk (VaR) Extreme Value Theory (EVT) References

More information

Statistics Introductory Correlation

Statistics Introductory Correlation Statistics Introductory Correlation Session 10 oscardavid.barrerarodriguez@sciencespo.fr April 9, 2018 Outline 1 Statistics are not used only to describe central tendency and variability for a single variable.

More information

Asymptotic Statistics-VI. Changliang Zou

Asymptotic Statistics-VI. Changliang Zou Asymptotic Statistics-VI Changliang Zou Kolmogorov-Smirnov distance Example (Kolmogorov-Smirnov confidence intervals) We know given α (0, 1), there is a well-defined d = d α,n such that, for any continuous

More information

Nonlinear Bivariate Comovements of Asset Prices: Theory and Tests

Nonlinear Bivariate Comovements of Asset Prices: Theory and Tests Nonlinear Bivariate Comovements of Asset Prices: Theory and Tests M. Corazza, A.G. Malliaris, E. Scalco Department of Applied Mathematics University Ca Foscari of Venice (Italy) Department of Economics

More information

Normal Probability Plot Probability Probability

Normal Probability Plot Probability Probability Modelling multivariate returns Stefano Herzel Department ofeconomics, University of Perugia 1 Catalin Starica Department of Mathematical Statistics, Chalmers University of Technology Reha Tutuncu Department

More information

FINM6900 Finance Theory Noisy Rational Expectations Equilibrium for Multiple Risky Assets

FINM6900 Finance Theory Noisy Rational Expectations Equilibrium for Multiple Risky Assets FINM69 Finance Theory Noisy Rational Expectations Equilibrium for Multiple Risky Assets February 3, 212 Reference Anat R. Admati, A Noisy Rational Expectations Equilibrium for Multi-Asset Securities Markets,

More information

Quantitative Introduction ro Risk and Uncertainty in Business Module 5: Hypothesis Testing

Quantitative Introduction ro Risk and Uncertainty in Business Module 5: Hypothesis Testing Quantitative Introduction ro Risk and Uncertainty in Business Module 5: Hypothesis Testing M. Vidyasagar Cecil & Ida Green Chair The University of Texas at Dallas Email: M.Vidyasagar@utdallas.edu October

More information

arxiv: v1 [math.pr] 24 Sep 2018

arxiv: v1 [math.pr] 24 Sep 2018 A short note on Anticipative portfolio optimization B. D Auria a,b,1,, J.-A. Salmerón a,1 a Dpto. Estadística, Universidad Carlos III de Madrid. Avda. de la Universidad 3, 8911, Leganés (Madrid Spain b

More information

Explicit Bounds for the Distribution Function of the Sum of Dependent Normally Distributed Random Variables

Explicit Bounds for the Distribution Function of the Sum of Dependent Normally Distributed Random Variables Explicit Bounds for the Distribution Function of the Sum of Dependent Normally Distributed Random Variables Walter Schneider July 26, 20 Abstract In this paper an analytic expression is given for the bounds

More information

Consistent tests for risk seeking behavior: A stochastic dominance approach

Consistent tests for risk seeking behavior: A stochastic dominance approach AHENS UNIVERSIY OF ECONOMICS AND BUSINESS DEPARMEN OF ECONOMICS WORKING PAPER SERIES 11-2015 Consistent tests for risk seeking behavior: A stochastic dominance approach Stelios Arvanitis Nikolas opaloglou

More information

SPECIFICATION TESTS IN PARAMETRIC VALUE-AT-RISK MODELS

SPECIFICATION TESTS IN PARAMETRIC VALUE-AT-RISK MODELS SPECIFICATION TESTS IN PARAMETRIC VALUE-AT-RISK MODELS J. Carlos Escanciano Indiana University, Bloomington, IN, USA Jose Olmo City University, London, UK Abstract One of the implications of the creation

More information

Does k-th Moment Exist?

Does k-th Moment Exist? Does k-th Moment Exist? Hitomi, K. 1 and Y. Nishiyama 2 1 Kyoto Institute of Technology, Japan 2 Institute of Economic Research, Kyoto University, Japan Email: hitomi@kit.ac.jp Keywords: Existence of moments,

More information

FE670 Algorithmic Trading Strategies. Stevens Institute of Technology

FE670 Algorithmic Trading Strategies. Stevens Institute of Technology FE670 Algorithmic Trading Strategies Lecture 8. Robust Portfolio Optimization Steve Yang Stevens Institute of Technology 10/17/2013 Outline 1 Robust Mean-Variance Formulations 2 Uncertain in Expected Return

More information

Political Science 236 Hypothesis Testing: Review and Bootstrapping

Political Science 236 Hypothesis Testing: Review and Bootstrapping Political Science 236 Hypothesis Testing: Review and Bootstrapping Rocío Titiunik Fall 2007 1 Hypothesis Testing Definition 1.1 Hypothesis. A hypothesis is a statement about a population parameter The

More information

1: PROBABILITY REVIEW

1: PROBABILITY REVIEW 1: PROBABILITY REVIEW Marek Rutkowski School of Mathematics and Statistics University of Sydney Semester 2, 2016 M. Rutkowski (USydney) Slides 1: Probability Review 1 / 56 Outline We will review the following

More information

Exact goodness-of-fit tests for censored data

Exact goodness-of-fit tests for censored data Exact goodness-of-fit tests for censored data Aurea Grané Statistics Department. Universidad Carlos III de Madrid. Abstract The statistic introduced in Fortiana and Grané (23, Journal of the Royal Statistical

More information

Higher order moments of the estimated tangency portfolio weights

Higher order moments of the estimated tangency portfolio weights WORKING PAPER 0/07 Higher order moments of the estimated tangency portfolio weights Farrukh Javed, Stepan Mazur, Edward Ngailo Statistics ISSN 403-0586 http://www.oru.se/institutioner/handelshogskolan-vid-orebro-universitet/forskning/publikationer/working-papers/

More information

Bootstrap tests of multiple inequality restrictions on variance ratios

Bootstrap tests of multiple inequality restrictions on variance ratios Economics Letters 91 (2006) 343 348 www.elsevier.com/locate/econbase Bootstrap tests of multiple inequality restrictions on variance ratios Jeff Fleming a, Chris Kirby b, *, Barbara Ostdiek a a Jones Graduate

More information

Bivariate Paired Numerical Data

Bivariate Paired Numerical Data Bivariate Paired Numerical Data Pearson s correlation, Spearman s ρ and Kendall s τ, tests of independence University of California, San Diego Instructor: Ery Arias-Castro http://math.ucsd.edu/~eariasca/teaching.html

More information

A Correction. Joel Peress INSEAD. Abstract

A Correction. Joel Peress INSEAD. Abstract Wealth, Information Acquisition and ortfolio Choice A Correction Joel eress INSEAD Abstract There is an error in my 2004 paper Wealth, Information Acquisition and ortfolio Choice. This note shows how to

More information

Dynamic Matrix-Variate Graphical Models A Synopsis 1

Dynamic Matrix-Variate Graphical Models A Synopsis 1 Proc. Valencia / ISBA 8th World Meeting on Bayesian Statistics Benidorm (Alicante, Spain), June 1st 6th, 2006 Dynamic Matrix-Variate Graphical Models A Synopsis 1 Carlos M. Carvalho & Mike West ISDS, Duke

More information

Consistent Tests for Conditional Treatment Effects

Consistent Tests for Conditional Treatment Effects Consistent Tests for Conditional Treatment Effects Yu-Chin Hsu Department of Economics University of Missouri at Columbia Preliminary: please do not cite or quote without permission.) This version: May

More information

Probabilities & Statistics Revision

Probabilities & Statistics Revision Probabilities & Statistics Revision Christopher Ting Christopher Ting http://www.mysmu.edu/faculty/christophert/ : christopherting@smu.edu.sg : 6828 0364 : LKCSB 5036 January 6, 2017 Christopher Ting QF

More information

Generalized Cramér von Mises goodness-of-fit tests for multivariate distributions

Generalized Cramér von Mises goodness-of-fit tests for multivariate distributions Hong Kong Baptist University HKBU Institutional Repository HKBU Staff Publication 009 Generalized Cramér von Mises goodness-of-fit tests for multivariate distributions Sung Nok Chiu Hong Kong Baptist University,

More information

Exact goodness-of-fit tests for censored data

Exact goodness-of-fit tests for censored data Ann Inst Stat Math ) 64:87 3 DOI.7/s463--356-y Exact goodness-of-fit tests for censored data Aurea Grané Received: February / Revised: 5 November / Published online: 7 April The Institute of Statistical

More information

NCoVaR Granger Causality

NCoVaR Granger Causality NCoVaR Granger Causality Cees Diks 1 Marcin Wolski 2 1 Universiteit van Amsterdam 2 European Investment Bank Bank of Italy Rome, 26 January 2018 The opinions expressed herein are those of the authors and

More information

Asymmetric Dependence, Tail Dependence, and the. Time Interval over which the Variables Are Measured

Asymmetric Dependence, Tail Dependence, and the. Time Interval over which the Variables Are Measured Asymmetric Dependence, Tail Dependence, and the Time Interval over which the Variables Are Measured Byoung Uk Kang and Gunky Kim Preliminary version: August 30, 2013 Comments Welcome! Kang, byoung.kang@polyu.edu.hk,

More information

Econ 423 Lecture Notes: Additional Topics in Time Series 1

Econ 423 Lecture Notes: Additional Topics in Time Series 1 Econ 423 Lecture Notes: Additional Topics in Time Series 1 John C. Chao April 25, 2017 1 These notes are based in large part on Chapter 16 of Stock and Watson (2011). They are for instructional purposes

More information

Interpreting Regression Results

Interpreting Regression Results Interpreting Regression Results Carlo Favero Favero () Interpreting Regression Results 1 / 42 Interpreting Regression Results Interpreting regression results is not a simple exercise. We propose to split

More information

Asymptotic distribution of the sample average value-at-risk

Asymptotic distribution of the sample average value-at-risk Asymptotic distribution of the sample average value-at-risk Stoyan V. Stoyanov Svetlozar T. Rachev September 3, 7 Abstract In this paper, we prove a result for the asymptotic distribution of the sample

More information

Summary: the confidence interval for the mean (σ 2 known) with gaussian assumption

Summary: the confidence interval for the mean (σ 2 known) with gaussian assumption Summary: the confidence interval for the mean (σ known) with gaussian assumption on X Let X be a Gaussian r.v. with mean µ and variance σ. If X 1, X,..., X n is a random sample drawn from X then the confidence

More information

STAT 461/561- Assignments, Year 2015

STAT 461/561- Assignments, Year 2015 STAT 461/561- Assignments, Year 2015 This is the second set of assignment problems. When you hand in any problem, include the problem itself and its number. pdf are welcome. If so, use large fonts and

More information

Class 2 & 3 Overfitting & Regularization

Class 2 & 3 Overfitting & Regularization Class 2 & 3 Overfitting & Regularization Carlo Ciliberto Department of Computer Science, UCL October 18, 2017 Last Class The goal of Statistical Learning Theory is to find a good estimator f n : X Y, approximating

More information

Multivariate Asset Return Prediction with Mixture Models

Multivariate Asset Return Prediction with Mixture Models Multivariate Asset Return Prediction with Mixture Models Swiss Banking Institute, University of Zürich Introduction The leptokurtic nature of asset returns has spawned an enormous amount of research into

More information

Aggregate Risk. MFM Practitioner Module: Quantitative Risk Management. John Dodson. February 6, Aggregate Risk. John Dodson.

Aggregate Risk. MFM Practitioner Module: Quantitative Risk Management. John Dodson. February 6, Aggregate Risk. John Dodson. MFM Practitioner Module: Quantitative Risk Management February 6, 2019 As we discussed last semester, the general goal of risk measurement is to come up with a single metric that can be used to make financial

More information

Expected Shortfall is not elicitable so what?

Expected Shortfall is not elicitable so what? Expected Shortfall is not elicitable so what? Dirk Tasche Bank of England Prudential Regulation Authority 1 dirk.tasche@gmx.net Finance & Stochastics seminar Imperial College, November 20, 2013 1 The opinions

More information

Fundamentals in Optimal Investments. Lecture I

Fundamentals in Optimal Investments. Lecture I Fundamentals in Optimal Investments Lecture I + 1 Portfolio choice Portfolio allocations and their ordering Performance indices Fundamentals in optimal portfolio choice Expected utility theory and its

More information

Information Choice in Macroeconomics and Finance.

Information Choice in Macroeconomics and Finance. Information Choice in Macroeconomics and Finance. Laura Veldkamp New York University, Stern School of Business, CEPR and NBER Spring 2009 1 Veldkamp What information consumes is rather obvious: It consumes

More information

Estimating the accuracy of a hypothesis Setting. Assume a binary classification setting

Estimating the accuracy of a hypothesis Setting. Assume a binary classification setting Estimating the accuracy of a hypothesis Setting Assume a binary classification setting Assume input/output pairs (x, y) are sampled from an unknown probability distribution D = p(x, y) Train a binary classifier

More information

Performance Evaluation and Comparison

Performance Evaluation and Comparison Outline Hong Chang Institute of Computing Technology, Chinese Academy of Sciences Machine Learning Methods (Fall 2012) Outline Outline I 1 Introduction 2 Cross Validation and Resampling 3 Interval Estimation

More information

Asymptotic distribution of the sample average value-at-risk in the case of heavy-tailed returns

Asymptotic distribution of the sample average value-at-risk in the case of heavy-tailed returns Asymptotic distribution of the sample average value-at-risk in the case of heavy-tailed returns Stoyan V. Stoyanov Chief Financial Researcher, FinAnalytica Inc., Seattle, USA e-mail: stoyan.stoyanov@finanalytica.com

More information

Expecting the Unexpected: Uniform Quantile Regression Bands with an application to Investor Sentiments

Expecting the Unexpected: Uniform Quantile Regression Bands with an application to Investor Sentiments Expecting the Unexpected: Uniform Bands with an application to Investor Sentiments Boston University November 16, 2016 Econometric Analysis of Heterogeneity in Financial Markets Using s Chapter 1: Expecting

More information

Discrete-Time Finite-Horizon Optimal ALM Problem with Regime-Switching for DB Pension Plan

Discrete-Time Finite-Horizon Optimal ALM Problem with Regime-Switching for DB Pension Plan Applied Mathematical Sciences, Vol. 10, 2016, no. 33, 1643-1652 HIKARI Ltd, www.m-hikari.com http://dx.doi.org/10.12988/ams.2016.6383 Discrete-Time Finite-Horizon Optimal ALM Problem with Regime-Switching

More information

Birgit Rudloff Operations Research and Financial Engineering, Princeton University

Birgit Rudloff Operations Research and Financial Engineering, Princeton University TIME CONSISTENT RISK AVERSE DYNAMIC DECISION MODELS: AN ECONOMIC INTERPRETATION Birgit Rudloff Operations Research and Financial Engineering, Princeton University brudloff@princeton.edu Alexandre Street

More information

Multivariate Distributions

Multivariate Distributions IEOR E4602: Quantitative Risk Management Spring 2016 c 2016 by Martin Haugh Multivariate Distributions We will study multivariate distributions in these notes, focusing 1 in particular on multivariate

More information

MA 575 Linear Models: Cedric E. Ginestet, Boston University Non-parametric Inference, Polynomial Regression Week 9, Lecture 2

MA 575 Linear Models: Cedric E. Ginestet, Boston University Non-parametric Inference, Polynomial Regression Week 9, Lecture 2 MA 575 Linear Models: Cedric E. Ginestet, Boston University Non-parametric Inference, Polynomial Regression Week 9, Lecture 2 1 Bootstrapped Bias and CIs Given a multiple regression model with mean and

More information

Statistical inference on Lévy processes

Statistical inference on Lévy processes Alberto Coca Cabrero University of Cambridge - CCA Supervisors: Dr. Richard Nickl and Professor L.C.G.Rogers Funded by Fundación Mutua Madrileña and EPSRC MASDOC/CCA student workshop 2013 26th March Outline

More information

Robustní monitorování stability v modelu CAPM

Robustní monitorování stability v modelu CAPM Robustní monitorování stability v modelu CAPM Ondřej Chochola, Marie Hušková, Zuzana Prášková (MFF UK) Josef Steinebach (University of Cologne) ROBUST 2012, Němčičky, 10.-14.9. 2012 Contents Introduction

More information

SYSM 6303: Quantitative Introduction to Risk and Uncertainty in Business Lecture 4: Fitting Data to Distributions

SYSM 6303: Quantitative Introduction to Risk and Uncertainty in Business Lecture 4: Fitting Data to Distributions SYSM 6303: Quantitative Introduction to Risk and Uncertainty in Business Lecture 4: Fitting Data to Distributions M. Vidyasagar Cecil & Ida Green Chair The University of Texas at Dallas Email: M.Vidyasagar@utdallas.edu

More information

Extreme Value Theory.

Extreme Value Theory. Bank of England Centre for Central Banking Studies CEMLA 2013 Extreme Value Theory. David G. Barr November 21, 2013 Any views expressed are those of the author and not necessarily those of the Bank of

More information

Stat 710: Mathematical Statistics Lecture 31

Stat 710: Mathematical Statistics Lecture 31 Stat 710: Mathematical Statistics Lecture 31 Jun Shao Department of Statistics University of Wisconsin Madison, WI 53706, USA Jun Shao (UW-Madison) Stat 710, Lecture 31 April 13, 2009 1 / 13 Lecture 31:

More information

Perturbative Approaches for Robust Intertemporal Optimal Portfolio Selection

Perturbative Approaches for Robust Intertemporal Optimal Portfolio Selection Perturbative Approaches for Robust Intertemporal Optimal Portfolio Selection F. Trojani and P. Vanini ECAS Course, Lugano, October 7-13, 2001 1 Contents Introduction Merton s Model and Perturbative Solution

More information

A simple graphical method to explore tail-dependence in stock-return pairs

A simple graphical method to explore tail-dependence in stock-return pairs A simple graphical method to explore tail-dependence in stock-return pairs Klaus Abberger, University of Konstanz, Germany Abstract: For a bivariate data set the dependence structure can not only be measured

More information

Long-Run Covariability

Long-Run Covariability Long-Run Covariability Ulrich K. Müller and Mark W. Watson Princeton University October 2016 Motivation Study the long-run covariability/relationship between economic variables great ratios, long-run Phillips

More information

Notes on Recursive Utility. Consider the setting of consumption in infinite time under uncertainty as in

Notes on Recursive Utility. Consider the setting of consumption in infinite time under uncertainty as in Notes on Recursive Utility Consider the setting of consumption in infinite time under uncertainty as in Section 1 (or Chapter 29, LeRoy & Werner, 2nd Ed.) Let u st be the continuation utility at s t. That

More information

Complexity of two and multi-stage stochastic programming problems

Complexity of two and multi-stage stochastic programming problems Complexity of two and multi-stage stochastic programming problems A. Shapiro School of Industrial and Systems Engineering, Georgia Institute of Technology, Atlanta, Georgia 30332-0205, USA The concept

More information

Generalized Hypothesis Testing and Maximizing the Success Probability in Financial Markets

Generalized Hypothesis Testing and Maximizing the Success Probability in Financial Markets Generalized Hypothesis Testing and Maximizing the Success Probability in Financial Markets Tim Leung 1, Qingshuo Song 2, and Jie Yang 3 1 Columbia University, New York, USA; leung@ieor.columbia.edu 2 City

More information

Motivation General concept of CVaR Optimization Comparison. VaR and CVaR. Přemysl Bejda.

Motivation General concept of CVaR Optimization Comparison. VaR and CVaR. Přemysl Bejda. VaR and CVaR Přemysl Bejda premyslbejda@gmail.com 2014 Contents 1 Motivation 2 General concept of CVaR 3 Optimization 4 Comparison Contents 1 Motivation 2 General concept of CVaR 3 Optimization 4 Comparison

More information

The Analysis of Power for Some Chosen VaR Backtesting Procedures - Simulation Approach

The Analysis of Power for Some Chosen VaR Backtesting Procedures - Simulation Approach The Analysis of Power for Some Chosen VaR Backtesting Procedures - Simulation Approach Krzysztof Piontek Department of Financial Investments and Risk Management Wroclaw University of Economics ul. Komandorska

More information

Regression Analysis. y t = β 1 x t1 + β 2 x t2 + β k x tk + ϵ t, t = 1,..., T,

Regression Analysis. y t = β 1 x t1 + β 2 x t2 + β k x tk + ϵ t, t = 1,..., T, Regression Analysis The multiple linear regression model with k explanatory variables assumes that the tth observation of the dependent or endogenous variable y t is described by the linear relationship

More information

A measure of radial asymmetry for bivariate copulas based on Sobolev norm

A measure of radial asymmetry for bivariate copulas based on Sobolev norm A measure of radial asymmetry for bivariate copulas based on Sobolev norm Ahmad Alikhani-Vafa Ali Dolati Abstract The modified Sobolev norm is used to construct an index for measuring the degree of radial

More information

Practical Statistics

Practical Statistics Practical Statistics Lecture 1 (Nov. 9): - Correlation - Hypothesis Testing Lecture 2 (Nov. 16): - Error Estimation - Bayesian Analysis - Rejecting Outliers Lecture 3 (Nov. 18) - Monte Carlo Modeling -

More information

Expected Shortfall is not elicitable so what?

Expected Shortfall is not elicitable so what? Expected Shortfall is not elicitable so what? Dirk Tasche Bank of England Prudential Regulation Authority 1 dirk.tasche@gmx.net Modern Risk Management of Insurance Firms Hannover, January 23, 2014 1 The

More information

Multivariate GARCH models.

Multivariate GARCH models. Multivariate GARCH models. Financial market volatility moves together over time across assets and markets. Recognizing this commonality through a multivariate modeling framework leads to obvious gains

More information

Chapter 5 Confidence Intervals

Chapter 5 Confidence Intervals Chapter 5 Confidence Intervals Confidence Intervals about a Population Mean, σ, Known Abbas Motamedi Tennessee Tech University A point estimate: a single number, calculated from a set of data, that is

More information

Statistical Methods for Particle Physics Lecture 1: parameter estimation, statistical tests

Statistical Methods for Particle Physics Lecture 1: parameter estimation, statistical tests Statistical Methods for Particle Physics Lecture 1: parameter estimation, statistical tests http://benasque.org/2018tae/cgi-bin/talks/allprint.pl TAE 2018 Benasque, Spain 3-15 Sept 2018 Glen Cowan Physics

More information

Framework for Analyzing Spatial Contagion between Financial Markets

Framework for Analyzing Spatial Contagion between Financial Markets Finance Letters, 2004, 2 (6), 8-15 Framework for Analyzing Spatial Contagion between Financial Markets Brendan O. Bradley a and Murad S. Taqqu b, a Acadian Asset Management Inc., USA b Boston University,

More information

Fall 2017 STAT 532 Homework Peter Hoff. 1. Let P be a probability measure on a collection of sets A.

Fall 2017 STAT 532 Homework Peter Hoff. 1. Let P be a probability measure on a collection of sets A. 1. Let P be a probability measure on a collection of sets A. (a) For each n N, let H n be a set in A such that H n H n+1. Show that P (H n ) monotonically converges to P ( k=1 H k) as n. (b) For each n

More information

ORIGINS OF STOCHASTIC PROGRAMMING

ORIGINS OF STOCHASTIC PROGRAMMING ORIGINS OF STOCHASTIC PROGRAMMING Early 1950 s: in applications of Linear Programming unknown values of coefficients: demands, technological coefficients, yields, etc. QUOTATION Dantzig, Interfaces 20,1990

More information

Volatility. Gerald P. Dwyer. February Clemson University

Volatility. Gerald P. Dwyer. February Clemson University Volatility Gerald P. Dwyer Clemson University February 2016 Outline 1 Volatility Characteristics of Time Series Heteroskedasticity Simpler Estimation Strategies Exponentially Weighted Moving Average Use

More information

1 Description of variables

1 Description of variables 1 Description of variables We have three possible instruments/state variables: dividend yield d t+1, default spread y t+1, and realized market volatility v t+1 d t is the continuously compounded 12 month

More information

A SECOND ORDER STOCHASTIC DOMINANCE PORTFOLIO EFFICIENCY MEASURE

A SECOND ORDER STOCHASTIC DOMINANCE PORTFOLIO EFFICIENCY MEASURE K Y B E R N E I K A V O L U M E 4 4 ( 2 0 0 8 ), N U M B E R 2, P A G E S 2 4 3 2 5 8 A SECOND ORDER SOCHASIC DOMINANCE PORFOLIO EFFICIENCY MEASURE Miloš Kopa and Petr Chovanec In this paper, we introduce

More information

Uncertainty in energy system models

Uncertainty in energy system models Uncertainty in energy system models Amy Wilson Durham University May 2015 Table of Contents 1 Model uncertainty 2 3 Example - generation investment 4 Conclusion Model uncertainty Contents 1 Model uncertainty

More information

University of California Berkeley

University of California Berkeley Working Paper #2018-02 Infinite Horizon CCAPM with Stochastic Taxation and Monetary Policy Revised from the Center for Risk Management Research Working Paper 2018-01 Konstantin Magin, University of California,

More information

ECOM 009 Macroeconomics B. Lecture 2

ECOM 009 Macroeconomics B. Lecture 2 ECOM 009 Macroeconomics B Lecture 2 Giulio Fella c Giulio Fella, 2014 ECOM 009 Macroeconomics B - Lecture 2 40/197 Aim of consumption theory Consumption theory aims at explaining consumption/saving decisions

More information

Bayesian Semiparametric GARCH Models

Bayesian Semiparametric GARCH Models Bayesian Semiparametric GARCH Models Xibin (Bill) Zhang and Maxwell L. King Department of Econometrics and Business Statistics Faculty of Business and Economics xibin.zhang@monash.edu Quantitative Methods

More information

Identifying Financial Risk Factors

Identifying Financial Risk Factors Identifying Financial Risk Factors with a Low-Rank Sparse Decomposition Lisa Goldberg Alex Shkolnik Berkeley Columbia Meeting in Engineering and Statistics 24 March 2016 Outline 1 A Brief History of Factor

More information

Scenario estimation and generation

Scenario estimation and generation October 10, 2004 The one-period case Distances of Probability Measures Tensor products of trees Tree reduction A decision problem is subject to uncertainty Uncertainty is represented by probability To

More information

Bayesian Semiparametric GARCH Models

Bayesian Semiparametric GARCH Models Bayesian Semiparametric GARCH Models Xibin (Bill) Zhang and Maxwell L. King Department of Econometrics and Business Statistics Faculty of Business and Economics xibin.zhang@monash.edu Quantitative Methods

More information