An alternative route to performance hypothesis testing Received (in revised form): 7th November, 2003
|
|
- Harry Brooks
- 5 years ago
- Views:
Transcription
1 An alternative route to perforance hypothesis testing Received (in revised for): 7th Noveber, 3 Bernd Scherer heads Research for Deutsche Asset Manageent in Europe. Before joining Deutsche, he worked at Morgan Stanley, J.P. Morgan and Schroders, where he headed global fixed incoe research. He has published several books on asset anageent and is a lecturer in Finance at the University of Augsburg. Head of Investent Solutions, Deutsche Asset Manageent, Mainzer Landstr , 637 Frankfurt, Gerany Tel: ; Fax: ; e-ail: bernd.scherer@db.co Abstract A wide variety of risk return ratios are routinely reported in sales pitches as well as acadeic publications. Little attept has been ade, however, to look at the sall saple distributions of these estiators in order to derive confidence bands. The reason for this has been the extree difficulty of working out the required statistics for ost risk return ratios. Rather than following classical statistics, this paper relies on a general and robust ethod which not only provides confidence intervals for arbitrary risk return ratios, saple sizes and distribution, but is also fairly easy to ipleent. Keywords: Sharpe ratio, Sortino ratio, hedge funds, bootstrapping, confidence interval, sall saple Introduction Reported risk return ratios relate average returns to alternative easures of risk, and hence involve the ratio of a rando (owing to sapling error) noinator and denoinator. As such, point estiates of these ratios are easy to calculate, but confidence intervals are uch ore difficult to arrive at. Confidence intervals are needed, however, for any kind of statistical inference and decision aking. While Lo () has shown that an asyptotic distribution exists for the Sharpe ratio, this result provides a special case rather than a general concept. Traditional ethods only work well if the sapling distribution of the statistic in question is asyptotically norally distributed. There is little guidance on the sall saple behaviour of risk adjusted perforance easures or how any data points one needs to rely on for asyptotic results. Useful exceptions are Pedersen and Satchell (), Miller and Gehr (1978) and Jobson and Korkie (1981). Moreover, these analytical solutions are either extreely difficult to work out or siply do not exist for odifications of the popular Sharpe ratio which focus ore on downside risk. An exaple is the well-known Sortino ratio (it relates average return to standard deviation of downside returns). A general ethod is needed which provides confidence intervals for arbitrary risk return ratios, saple sizes and distributions. Henry Stewart Publications X () Vol. 5, 1, 5 1 Journal of Asset Manageent 5
2 Scherer Bootstrapping theory as an alternative Suppose a series of returns (total return inus risk-free rate) r 1, r,..., r is observed. Ex post risk return ratios ( ˆ) are calculated as the ratio of average return per unit of risk. This paper will focus on Sharpe and Sortino ratios (Sharpe, 199; Sortino and Price, 199). Both ratios differ with respect to the eployed risk easure. The Sharpe ratio applies a syetric risk concept, equally penalising (squaring) downside and upside deviations fro the saple ean return, while the Sortino ratio only includes negative perforance in their calculation of squared returns. The Sortino ratio is included for three reasons. First, it provides a better capture of risk if returns are non-norally distributed, as is the case for hedge fund returns. Secondly, it is known to suffer ore fro estiation error, as it uses fewer data (extree returns are by definition rare). Thirdly, no large saple approxiations exist. The saple calculations for both ratios are given below. Sharpe ratio Sortino ratio 1 1 r i i=1 i=1 (r i r) (1) r i i=1 i=1 I(r i <)(r i ) () where I(r i < ) denotes the indicator function. High ratios are preferable (everythingelseequal)astheyindicatea better return per unit of risk taken. If the sall saple distribution of ˆ is far fro noral, classical ethods are biased and unreliable. In any case, the analytical forula for the large sapling distribution of (1) is extreely hard to coe by, while it is unknown for (). In order to overcoe the above proble, bootstrapping techniques are relied on. Resapling treats the current saple as a good approxiation of the true distribution (in the absence of further inforation, it is the best available). It then repeatedly draws fro the epirical distribution to recalculate the statistic of interest any ties to arrive at the bootstrap sapling distribution that can now be used for hypothesis testing. Suppose one is given onthly returns on the HFR fund of funds index ranging fro January 199 to April 3 (16 observations). The JPM one-onth cash rate fro DataStrea is used to calculate a risk-free return. The bootstrapping procedure is as follows. Randoly draw 16 (original saple size) returns with replaceent fro the original saple. Calculate a new risk return ratio ˆ*b based on the resapled returns. Repeat this procedure for b 1,..., B ties arriving at ˆ*1, ˆ*1, L, ˆ*b,L ˆ*B resapled ratios. The bootstrap sapling distribution of ˆ*b can now be taken to judge whether the sapling distribution ˆ of in sall saples is noral (and hence traditional approaches ight not be so bad after all) or not. With B 1,, the following results are obtained (see Figure 1). 1 While the Sharpe ratio is well behaved (all resapled realisations plot on a straight line, equalising hypothetical and epirical percentiles), the sae cannot be said about the Sortino ratio. Deviations fro norality are large at both ends. Noral (sall saple) approxiations look only reasonable for the traditional Sharpe ratio. They see to be largely isleading, however, for the Sortino ratio. Taking the.5 per cent and 97.5 per cent percentile, one can now arrive at a syetric 95 per cent confidence band CI( ˆ*.5%, ˆ*97.5%). Note the apparent non-norality of the saple distribution for the Sortino ratio in 6 Journal of Asset Manageent Vol. 5, 1, 5 1 Henry Stewart Publications X ()
3 Perforance hypothesis testing SORTINO RATIO SHARPE RATIO Quantiles of Standard Noral Quantiles of Standard Noral Figure 1 QQ-plot for bootstrapped sapling distribution Figure Bootstrapped sapling distributions Henry Stewart Publications X () Vol. 5, 1, 5 1 Journal of Asset Manageent 7
4 Scherer Probability that double bootstrapped Sortino-ratio falls below initial Sortino-ratio Figure 3 Saple histogra of u b 1 Z Z z=1i( ˆ *b * z< ˆ) Figure. As suspected, the Sortino ratio shows a uch larger dispersion in resapled outcoes and hence estiation error. The corresponding values can be cut off fro the histogras in Figure. For the Sortino (Sharpe) ratio, the 95 per cent interval ranges fro.11 to.9 (.8.1). Both ratios are significantly different fro zero. Increasing the coverage ratio with the double bootstrap So far, reliance has been on the 95 per cent interval fro a siple bootstrap procedure. It is not guaranteed, however, that the 95 per cent confidence band calculated above indeed covers the true ratio with 95 per cent probability. The true ratio ight only fall 85 per cent of all ties into the estiated confidence band. One ethod of increasing the coverage probability is the double bootstrap, which can be thought of as bootstrapping the bootstrap. The double bootstrap as described by Nankervis () involves the following calculations. Perfor the siple bootstrap as described above. Save all b 1,..., B resapled data sets as well as resapled ratios ˆ*b. Thisiscalledthefirst stage resapling. For each of the B resapled datasets, start a second round of z 1,..., Z resaples, leading to a total of BZ resaples denoted as ˆ*b * z.foreach ˆ*b, there exists a new set of Z resapled ratios ˆ*b * 1,L, ˆ*b * Z. These are the second stage resaples. For each ˆ*b, calculate the percentage of second stage resaples ˆ*b * 1,L, ˆ*b * Z that fall below the original saple estiate of the risk return ratio ˆ, ie calculate u b 1 Z Z z=1 I( ˆ*b * z < ˆ) (3) Choose B 1, and Z. Under ideal conditions u b follows a unifor distribution. Figure 3 shows that this assuption is clearly violated for the double bootstrapped Sortino ratios ˆ*b * 1,L, ˆ*b * Z. Foral tests such as the Kologorov 8 Journal of Asset Manageent Vol. 5, 1, 5 1 Henry Stewart Publications X ()
5 Perforance hypothesis testing (a) RAW DATA (b) FILTERED DATA ACF. ACF Lag Lag Figure Autocorrelation for hedge fund series and Sirnov test as well as the adjustent test provide p values close to per cent. Hence, the null hypothesis that Figure 3 coes fro a unifor distribution can be safely rejected. Finally, calculate the.5 per cent and 97.5 per cent percentiles of u b.usethese value to adjust the first stage resaple confidence band to CI( ˆ*u.5%, ˆ*u97.5%). With the double bootstrap, the confidence interval changes to a uch higher bound of.18 (instead of.11) representing the 8 per cent quantile (rather than the.5 per cent quantile). In effect, one gets CI( ˆ*u7.9%.18, ˆ*u96%.96). What if one needs to deal with autocorrelated data? So far, it has been iplicitly assued that return data r 1, r,..., r have been drawn independently, ie neither returns nor volatilities are autocorrelated. Bootstrapping iplicitly reoves any tie dependence by its very definition. All draws are unconditional on the result of the previous draw. After a strong positive draw, there is no echanis that would favour another positive draw (in case of positive autocorrelation) or another large draw (in case of ARCH effects). Many financial tie series (real estate, corporate bonds, high yield, hedge funds), however, show strong autocorrelation due to the illiquidity (infrequent trading) of the underlying instruents. Bootstrapping fails if one does not account for return dependence. For the data analysed so far, the autocorrelation function can be plotted as in Figure (a). The first-order autocorrelation coefficient aounts to.31 and is well outside the confidence liits (given by dotted line), ie it is statistically significant. Henry Stewart Publications X () Vol. 5, 1, 5 1 Journal of Asset Manageent 9
6 Scherer FREQUENCY 8 6 FREQUENCY LOWER BOUND (.5% LEVEL) FOR RAW DATA LOWER BOUND (.5% LEVEL) FOR FILTERED DATA Figure 5 Lower confidence bound (.5 per cent) on Sharpe ratio for raw and adjusted data One way to deal with this issue is to reove the autocorrelation using a siple filter that adjusts for the inherent AR(1) process (lag one is statistically significant, while lag two is not) and reapply the ethods established in the previous sections to the transfored series. It is known that the first-order autocorrelation can be reoved. Estiate the AR(1) coefficient fro a linear regression r t 1 r t 1 t () Save the regression coefficient 1 and create a new filtered series r t * according to r t * r t 1 1 r t 1 (5) Obviously this represents a break with the non-paraetric approach used so far. As a coproise, one ight want to estiate Equation () using robust ethods. The above procedure will leave the average return of a series constant, but effectively increase (decrease) its variance for positive autocorrelation (negative correlation) as described in Scherer (). The effect of this procedure can be seen in the right-hand autocorrelation plot in Figure (b). Virtually all significant autocorrelation (confidence bounds are given by dotted lines) has been reoved. After the reoval of autocorrelation, one can proceed as in the previous section. Rather than focusing on a single series, all HFR series are used for the described frequency and data period (see the HFR website for a description of the underlying data series). Figure 5 plots the distribution of lower confidence bounds (.5 per cent percentile) of Sharpe ratios for raw data as well as adjusted (filtered) data. The Sharpe ratios of 5 out of 37 series 1 Journal of Asset Manageent Vol. 5, 1, 5 1 Henry Stewart Publications X ()
7 Perforance hypothesis testing FREQUENCY 6 FREQUENCY LOWER BOUND (.5% LEVEL) FOR RAW DATA LOWER BOUND (.5% LEVEL) FOR FILTERED DATA Figure 6 Lower confidence bound (.5 per cent) on Sortino ratio for raw and adjusted data (expected:.5% 37 1) are not statistically different fro zero. This nuber increases to 11, if filtered data (with higher volatility) are used. If this is cobined with a suspicion of upward bias in hedge fund returns, any hedge fund styles fail to show statistically significant Sharpe ratios of even onthly returns. An alost identical picture can be found for the Sortino ratio in Figure 6. Again, the historical track record of the hedge fund industry as a whole leaves soe doubt about the industry s ability to create added value. Suary Renewed interest in the significance of risk return ratios has been focusing on closed for solutions for the well-known Sharpe ratio. This paper provided a robust ethodology for evaluating the properties of the Sharpe ratio s sapling distribution, as well as how to derive confidence intervals without having to rely on asyptotic approxiations (as in reality saples are sall). It has also been shown that the sapling distribution of other risk return easures for which asyptotic results do not exist ight be highly non-noral. While the double bootstrap ethodology leads to significantly refined confidence bands, a ore realistic odelling of hedge fund returns leaves the case for hedge funds less optiistic than providers of these services ight wish. Notes 1 All calculations have been perfored in S-Plus. For relevant code and further exaples on the use of bootstrapping in portfolio anageent see Scherer and Martin (3). References Jobson, J. and Korkie, B. (1981) Perforance Hypothesis Testing with the Sharpe and Treynor Measures, Journal of Finance, 36, Lo, A. W. () The Statistics of Sharpe Ratios, Financial Analysts Journal, July/August, 58 (). Henry Stewart Publications X () Vol. 5, 1, 5 1 Journal of Asset Manageent 11
8 Scherer Miller, R. and Gehr, A. (1978) Saple Bias and Perforance Measures: A Note, Journal of Financial and Quantitative Analysis, 13, Nankervis, J. () Stopping Rules for Double Bootstrap Confidence Intervals, University of Surrey, /Nankervis_John.pdf Pedersen, C. and Satchell, S. () Sall Saple Measures of Perforance Measures in the Asyetric Response Model, Journal of Financial and Quantitative Analysis, 35(3), 5 5. Scherer, B. () Portfolio Construction and Risk Budgeting, Riskwater, London. Scherer, B. and Martin, D. (3) Portfolio Optization using Nuopt for S-Plus, Springer, New York. Sharpe, W. F. (199) TheSharpeRatio, Journal of Portfolio Manageent, Fall, Sortino, F. A. and Price, L. N. (199) Perforance Measureent in a Downside Risk Fraework, Journal of Investing, Fall. 1 Journal of Asset Manageent Vol. 5, 1, 5 1 Henry Stewart Publications X ()
III. Valuation Framework for CDS options
III. Valuation Fraework for CDS options In siulation, the underlying asset price is the ost iportant variable. The suitable dynaics is selected to describe the underlying spreads. The relevant paraeters
More informationMAT 3788 Lecture 3, Feb
The Tie Value of Money MAT 3788 Lecture 3, Feb 010 The Tie Value of Money and Interest Rates Prof. Boyan Kostadinov, City Tech of CUNY Everyone is failiar with the saying "tie is oney" and in finance there
More informationTime Value of Money. Financial Mathematics for Actuaries Downloaded from by on 01/12/18. For personal use only.
Interest Accuulation and Tie Value of Money Fro tie to tie we are faced with probles of aking financial decisions. These ay involve anything fro borrowing a loan fro a bank to purchase a house or a car;
More informationFinancial Risk: Credit Risk, Lecture 1
Financial Risk: Credit Risk, Lecture 1 Alexander Herbertsson Centre For Finance/Departent of Econoics School of Econoics, Business and Law, University of Gothenburg E-ail: alexander.herbertsson@cff.gu.se
More informationA NUMERICAL EXAMPLE FOR PORTFOLIO OPTIMIZATION. In 2003, I collected data on 20 stocks, which are listed below: Berkshire-Hathaway B. Citigroup, Inc.
A NUMERICAL EXAMPLE FOR PORTFOLIO OPTIMIZATION In 3, I collected data on stocks, which are listed below: Sybol ADBE AMZN BA BRKB C CAT CSCO DD FDX GE GLW GM INTC JNJ KO MO MSFT RTN SBC Nae Adobe Systes
More informationAnalysis of the purchase option of computers
Analysis of the of coputers N. Ahituv and I. Borovits Faculty of Manageent, The Leon Recanati Graduate School of Business Adinistration, Tel-Aviv University, University Capus, Raat-Aviv, Tel-Aviv, Israel
More informationOptimal Design Of English Auctions With Discrete Bid Levels*
Optial Design Of English Auctions With Discrete Bid Levels* E. David, A. Rogers and N. R. Jennings Electronics and Coputer Science, University of Southapton, Southapton, SO7 BJ, UK. {ed,acr,nrj}@ecs.soton.ac.uk.
More information... About Higher Moments
WHAT PRACTITIONERS NEED TO KNOW...... About Higher Moents Mark P. Kritzan In financial analysis, a return distribution is coonly described by its expected return and standard deviation. For exaple, the
More informationCHAPTER 2: FUTURES MARKETS AND THE USE OF FUTURES FOR HEDGING
CHAPER : FUURES MARKES AND HE USE OF FUURES FOR HEDGING Futures contracts are agreeents to buy or sell an asset in the future for a certain price. Unlike forward contracts, they are usually traded on an
More information1. PAY $1: GET $2 N IF 1ST HEADS COMES UP ON NTH TOSS
APPLIED ECONOICS FOR ANAGERS SESSION I. REVIEW: EXTERNALITIES AND PUBLIC GOODS A. PROBLE IS ABSENCE OF PROPERTY RIGHTS B. REINTRODUCTION OF ARKET/PRICE ECHANIS C. PUBLIC GOODS AND TAXATION II. INFORATION
More informationIntroduction to Risk, Return and the Opportunity Cost of Capital
Introduction to Risk, Return and the Opportunity Cost of Capital Alexander Krüger, 008-09-30 Definitions and Forulas Investent risk There are three basic questions arising when we start thinking about
More informationSection on Survey Research Methods
Using the Statistics of Incoe Division s Saple Data to Reduce Measureent and Processing Error in Sall Area Estiates Produced fro Adinistrative Tax Records Kiberly Henry, Partha Lahiri, and Robin Fisher
More informationCapital Asset Pricing Model: The Criticisms and the Status Quo
Journal of Applied Sciences Research, 7(1): 33-41, 2011 ISSN 1819-544X This is a refereed journal and all articles are professionally screened and reviewed 33 ORIGINAL ARTICLES Capital Asset Pricing Model:
More informationUNCOVERED INTEREST PARITY IN CENTRAL AND EASTERN EUROPE: CONVERGENCE AND THE GLOBAL FINANCIAL CRISIS 1
UNCOVERED INTEREST PARITY IN CENTRAL AND EASTERN EUROPE: CONVERGENCE AND THE GLOBAL FINANCIAL CRISIS 1 Abstract Fabio Filipozzi 2, Karsten Staehr Tallinn University of Technology, Bank of Estonia This
More informationSurvey of Math: Chapter 21: Consumer Finance Savings Page 1
Survey of Math: Chapter 21: Consuer Finance Savings Page 1 The atheatical concepts we use to describe finance are also used to describe how populations of organiss vary over tie, how disease spreads through
More informationThe Institute of Chartered Accountants of Sri Lanka
The Institute of Chartered Accountants of Sri Lanka Executive Diploa in Business and Accounting Financial Matheatics Financial Matheatics deals with probles of investing Money, or Capital. If the investor
More informationAppendix Table A1. MPC Stratified by Additional Variables
Appendix Table A1. MPC Stratified by Additional Variables This table presents estiates of the MPC out of liquidity for groups of consuers stratified by whether they have low, ediu, or high levels of credit
More informationWho Gains and Who Loses from the 2011 Debit Card Interchange Fee Reform?
No. 12-6 Who Gains and Who Loses fro the 2011 Debit Card Interchange Fee Refor? Abstract: Oz Shy In October 2011, new rules governing debit card interchange fees becae effective in the United States. These
More informationVariance Swaps and Non-Constant Vega
Variance Swaps and Non-Constant Vega David E. Kuenzi Head of Risk anageent and Quantitative Research Glenwood Capital Investents, LLC 3 N. Wacker Drive, Suite 8 Chicago, IL 666 dkuenzi@glenwood.co Phone
More informationCONDITIONAL MEAN DOMINANCE: TESTING FOR SUFFICIENCY OF ANOMALIES
CONDITIONAL MEAN DOMINANCE: TESTING FOR SUFFICIENCY OF ANOMALIES K. Victor Chow and Ou Hu* ABSTRACT Extensive epirical literature of anoalies suggests that an asset reallocation by buying a subset of the
More informationResearch on the Management Strategy from the Perspective of Profit and Loss Balance
ISSN: 2278-3369 International Journal of Advances in Manageent and Econoics Available online at: www.anageentjournal.info RESEARCH ARTICLE Research on the Manageent Strategy fro the Perspective of Profit
More informationDepartment of Econometrics and Business Statistics
ISSN 440-77X Australia Departent of Econoetrics and Business Statistics http://www.buseco.onash.edu.au/depts/ebs/pubs/wpapers/ Applications of Inforation Measures to Assess Convergence in the Central Liit
More informationOn the population median estimation using quartile double ranked set sampling
On te population edian estiation using quartile double ranked set sapling Aer Ibrai Al-Oari Faculty of Science Departent of Mateatics Al al-bayt University Mafraq Jordan aloari_aer@yaoo.co Loai M. Al-Zubi
More informationGarrison Schlauch - CLAS. This handout covers every type of utility function you will see in Econ 10A.
This handout covers every type of utility function you will see in Econ 0A. Budget Constraint Unfortunately, we don t have unliited oney, and things cost oney. To siplify our analysis of constrained utility
More informationResearch Article Analysis on the Impact of the Fluctuation of the International Gold Prices on the Chinese Gold Stocks
Discrete Dynaics in Nature and Society, Article ID 308626, 6 pages http://dx.doi.org/10.1155/2014/308626 Research Article Analysis on the Ipact of the Fluctuation of the International Gold Prices on the
More informationRealized Variance and IID Market Microstructure Noise
Realized Variance and IID Market Microstructure Noise Peter R. Hansen a, Asger Lunde b a Brown University, Departent of Econoics, Box B,Providence, RI 02912, USA b Aarhus School of Business, Departent
More informationA Description of Swedish Producer and Import Price Indices PPI, EXPI and IMPI
STATSTCS SWEDE Rev. 2010-12-20 1(10) A Description of Swedish roducer and port rice ndices, EX and M The rice indices in roducer and port stages () ai to show the average change in prices in producer and
More informationTime Varying International Market Integration
International Journal of conoics and Finance; Vol. 5, No. 11; 013 ISSN 1916-971X-ISSN 1916-978 Published by Canadian Center of Science and ducation Tie Varying International Market Integration Dhouha Hadidane
More informationConstruction Methods.. Ch.-2- Factors Affecting the Selection of Construction Equipment
Construction Methods.. Ch.-2- Factors Affecting the Selection of Construction Equipent Chapter 2 Factors Affecting the Selection of Construction Equipent 2. Factors Affecting the Selection of Construction
More informationPRODUCTION COSTS MANAGEMENT BY MEANS OF INDIRECT COST ALLOCATED MODEL
PRODUCTION COSTS MANAGEMENT BY MEANS OF INDIRECT COST ALLOCATED MODEL Berislav Bolfek 1, Jasna Vujčić 2 1 Polytechnic Slavonski Brod, Croatia, berislav.bolfek@vusb.hr 2 High school ''Matija Antun Reljković'',
More informationAn Analytical Solution to Reasonable Royalty Rate Calculations a
-0- An Analytical Solution to Reasonable Royalty Rate Calculations a Willia Choi b Roy Weinstein c July 000 Abstract The courts are increasingly encouraging use of ore rigorous, scientific approaches to
More informationStaff Memo N O 2005/11. Documentation of the method used by Norges Bank for estimating implied forward interest rates.
N O 005/ Oslo Noveber 4, 005 Staff Meo Departent for Market Operations and Analysis Docuentation of the ethod used by Norges Bank for estiating iplied forward interest rates by Gaute Myklebust Publications
More informationEstimating Nonlinear Models With Multiply Imputed Data
Estiating onlinear Models With Multiply Iputed Data Catherine Phillips Montalto 1 and Yoonkyung Yuh 2 Repeated-iputation inference (RII) techniques for estiating nonlinear odels with ultiply iputed data
More informationHandelsbanken Debt Security Index Base Methodology. Version September 2017
Handelsbanken Debt Security Index Base ethodology Version 1.0 22 Septeber 2017 Contents 1 Introduction... 3 2 Description... 3 3 General Ters... 3 4 Iportant Inforation... 4 5 Definitions... 5 5.1 iscellaneous...
More informationS old. S new. Old p D. Old q. New q
Proble Set 1: Solutions ECON 301: Interediate Microeconoics Prof. Marek Weretka Proble 1 (Fro Varian Chapter 1) In this proble, the supply curve shifts to the left as soe of the apartents are converted
More informationHistorical Yield Curve Scenarios Generation without Resorting to Variance Reduction Techniques
Working Paper Series National Centre of Copetence in Research Financial Valuation and Risk Manageent Working Paper No. 136 Historical Yield Curve Scenarios Generation without Resorting to Variance Reduction
More informationIntroductory Financial Mathematics DSC1630
/2015 Tutorial Letter 201/1/2015 Introductory Financial Matheatics DSC1630 Seester 1 Departent of Decision Sciences Iportant Inforation: This tutorial letter contains the solutions of Assignent 01. Bar
More informationSee Market liquidity: Research Findings and Selected Policy Implications in BIS (1999) for the various dimensions of liquidity.
Estiating liquidity preia in the Spanish Governent securities arket 1 Francisco Alonso, Roberto Blanco, Ana del Río, Alicia Sanchís, Banco de España Abstract This paper investigates the presence of liquidity
More informationCalculating Value-at-Risk Using the Granularity Adjustment Method in the Portfolio Credit Risk Model with Random Loss Given Default
Journal of Econoics and Manageent, 016, Vol. 1, No., 157-176 Calculating Value-at-Risk Using the Granularity Adjustent Method in the Portfolio Credit Risk Model with Rando Loss Given Default Yi-Ping Chang
More informationASSESSING CREDIT LOSS DISTRIBUTIONS FOR INDIVIDUAL BORROWERS AND CREDIT PORTFOLIOS. BAYESIAN MULTI-PERIOD MODEL VS. BASEL II MODEL.
ASSESSING CREIT LOSS ISTRIBUTIONS FOR INIVIUAL BORROWERS AN CREIT PORTFOLIOS. BAYESIAN ULTI-PERIO OEL VS. BASEL II OEL. Leonid V. Philosophov,. Sc., Professor, oscow Coittee of Bankruptcy Affairs. 33 47
More informationNontradables and relative price levels across areas within Japan Hidehiro Ikeno Surugadai University
Nontradables and relative price levels across areas within Japan Hidehiro Ieno Surugadai University 1. Introduction This paper exaines epirically the iportance of tradables and nontradables in deterining
More informationAIM V.I. Small Cap Equity Fund
AIM V.I. Sall Cap Equity Fund PROSPECTUS May 1, 2009 Series I shares Shares of the fund are currently offered only to insurance copany separate accounts funding variable annuity contracts and variable
More informationState Trading Enterprises as Non-Tariff Measures: Theory, Evidence and Future Research Directions
State Trading Enterprises as Non-Tariff Measures: Theory, Evidence and Future Research Directions Steve McCorriston (University of Exeter, UK) (s.ccorriston@ex.ac.uk) Donald MacLaren (university of Melbourne,
More informationEarnings per share up by 21.2% to EUR 6.53, allowing increased dividend of EUR 2.40 per share
Press Release Mülhei an der Ruhr, March 21, 2013 Brenntag once again reports significantly iproved earnings and exceeds iddle of the guidance range Gross profit increased to 1,925.7 illion Growth in operating
More informationMonte Carlo Methods. Monte Carlo methods
ρ θ σ µ Monte Carlo Methos What is a Monte Carlo Metho? Rano walks The Metropolis rule iportance sapling Near neighbor sapling Sapling prior an posterior probability Exaple: gravity inversion The ovie
More informationEvaluation on the Growth of Listed SMEs Based on Improved Principal Component Projection Method
Proceedings of the 7th International Conference on Innovation & Manageent 519 Evaluation on the Growth of Listed SMEs Based on Iproved Principal Coponent Projection Method Li Li, Ci Jinfeng Shenzhen Graduate
More informationExtreme Risk Analysis July 2009
Extree Risk Analysis Lisa R. Goldberg Michael Y. Hayes Jose Menchero Indrajit Mitra Quantitative risk anageent allows for qualitative notions such as optiality and expected returns to be put on a quantitative
More informationDSC1630. Tutorial letter 201/1/2014. Introductory Financial Mathematics. Semester 1. Department of Decision Sciences DSC1630/201/1/2014
DSC1630/201/1/2014 Tutorial letter 201/1/2014 Introductory Financial Matheatics DSC1630 Seester 1 Departent of Decision Sciences IMPORTANT INFORMATION: This tutorial letter contains solutions to the assignents
More informationHiding Loan Losses: How to Do It? How to Eliminate It?
ömföäflsäafaäsflassflassf ffffffffffffffffffffffffffffffffffff Discussion Papers Hiding oan osses: How to Do It? How to Eliinate It? J P. Niiniäki Helsinki School of Econoics and HECER Discussion Paper
More informationCapital reserve planning:
C O - O P E R A T I V E H O U S I N G F E D E R A T I O N O F C A N A D A Capital reserve planning: A guide for federal-progra co-ops Getting our house in order P A R T O F T H E 2 0 2 0 V I S I O N T
More informationBond Duration. Floyd Vest
Bond Duration Floyd Vest It is well known that when arket interest rates change, the price of a bond, or the share prices in a bond fund, changes. Bond duration is widely used to estiate the change in
More information4. Martha S. has a choice of two assets: The first is a risk-free asset that offers a rate of return of r
Spring 009 010 / IA 350, Interediate Microeconoics / Proble Set 3 1. Suppose that a stock has a beta of 1.5, the return of the arket is 10%, and the risk-free rate of return is 5%. What is the expected
More informationWhy Do Large Investors Disclose Their Information?
Why Do Large Investors Disclose Their Inforation? Ying Liu Noveber 7, 2017 Abstract Large investors often advertise private inforation at private talks or in the edia. To analyse the incentives for inforation
More informationSTOCK PRICE AND EXCHANGE RATE CAUSALITY: THE CASE OF FOUR ASEAN COUNTRIES
Stock Price and Exchange Rate Causality: The Case of Four Asean Countries STOCK PRICE AND EXCHANGE RATE CAUSALITY: THE CASE OF FOUR ASEAN COUNTRIES D. Agus Harjito, Indonesian Islaic University Carl B.
More informationOPTIMIZATION APPROACHES IN RISK MANAGEMENT: APPLICATIONS IN FINANCE AND AGRICULTURE
OPTIMIZATION APPROACHES IN RISK MANAGEMENT: APPLICATIONS IN FINANCE AND AGRICULTURE By CHUNG-JUI WANG A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
More informationThe Least-Squares Method for American Option Pricing
U.U.D.M. Proect Report 29:6 The Least-Squares Method for Aerican Option Pricing Xueun Huang and Xuewen Huang Exaensarbete i ateatik, 3 hp + 5 hp Handledare och exainator: Macie Kliek Septeber 29 Departent
More informationBERMUDA NATIONAL PENSION SCHEME (GENERAL) REGULATIONS 1999 BR 82 / 1999
QUO FA T A F U E R N T BERMUDA NATIONAL PENSION SCHEME (GENERAL) REGULATIONS 1999 BR 82 / 1999 TABLE OF CONTENTS 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Citation Interpretation PART 1 PRELIMINARY PART II REGISTRATION
More informationNBER WORKING PAPER SERIES WEAK AND SEMI-STRONG FORM STOCK RETURN PREDICTABILITY, REVISITED. Wayne E. Ferson Andrea Heuson Tie Su
NBER WORKING PAPER SERIES WEAK AND SEMI-STRONG FORM STOCK RETURN PREDICTABILITY, REVISITED Wayne E. Ferson Andrea Heuson Tie Su Working Paper 10689 http://www.nber.org/papers/w10689 NATIONAL BUREAU OF
More informationSo What Do I Get? The Bank s View of Lending Relationships
So What Do I Get? The Bank s View of Lending Relationships Sreedhar Bharath, Sandeep Dahiya, Anthony Saunders, and Anand Srinivasan JEL Classification: G21; G24 Keywords: Lending relationships; Bank loans;
More informationExtremeVaR of South East Asian Stock Indices with Extreme Distribution- Based Efficiency
Available online at www.sciencedirect.co ScienceDirect Procedia Econoics and Finance 5 ( 2013 ) 120 124 International Conference on Applied Econoics (ICOAE) 2013 EtreeVaR of South East Asian Stock Indices
More informationPerformance Analysis of Online Anticipatory Algorithms for Large Multistage Stochastic Integer Programs
Perforance Analysis of Online Anticipatory Algoriths for Large Multistage Stochastic Integer Progras Luc Mercier and Pascal Van Hentenryck Brown University {ercier,pvh}@cs.brown.edu Abstract Despite significant
More informationUniversity of New South Wales Semester 1, Economics 4201 and Homework #2 Due on Tuesday 3/29 (20% penalty per day late)
University of New South Wales Semester 1, 2011 School of Economics James Morley 1. Autoregressive Processes (15 points) Economics 4201 and 6203 Homework #2 Due on Tuesday 3/29 (20 penalty per day late)
More informationAN ANALYSIS OF EQUITY IN INSURANCE. THE MATHEMATICAL APPROACH OF RISK OF RUIN FOR INSURERS
Iulian Mircea AN ANALYSIS OF EQUITY IN INSURANCE. THE MATHEMATICAL APPROACH OF RISK OF RUIN FOR INSURERS A.S.E. Bucure ti, CSIE, Str.Mihail Moxa nr. 5-7, irceaiulian9@yahoo.co, Tel.074.0.0.38 Paul T n
More informationNon-Linear Exchange Rate Pass-Through in Emerging Markets
WP/16/1 Non-Linear Exchange Rate Pass-Through in Eerging Markets by Francesca G. Caselli and Agustin Roitan IMF Working Papers describe research in progress by the author(s) and are published to elicit
More informationNeural Network Model of Pricing Health Care Insurance
Neural Network Model of Pricing Health Care Insurance Abstract To pricing health insurance plan statisticians use atheatical odels to analysis custoer s future health condition. General Addictive Model
More informationLecture 12: The Bootstrap
Lecture 12: The Bootstrap Reading: Chapter 5 STATS 202: Data mining and analysis October 20, 2017 1 / 16 Announcements Midterm is on Monday, Oct 30 Topics: chapters 1-5 and 10 of the book everything until
More informationTotal PS TG. Budgeted production levels can be calculated as follows:
U. ;' cn '.:. \.' >>.:---"--^ '-.'" * i--.'. * ::-;.v>"--:'i.-^ -7 -..=../.-' "-. " '.:.' Ill all it.;? s Solution Total PS TG Sales units 6,000 5,000 1,000 Sales value $605,000 $475,000 $130,000 Workings
More informationApproximating the Confidence Intervals for Sharpe Style Weights
Approximating the Confidence Intervals for Sharpe Style Weights Angelo Lobosco and Dan DiBartolomeo Style analysis is a form of constrained regression that uses a weighted combination of market indexes
More informationResearch on Entrepreneur Environment Management Evaluation Method Derived from Advantage Structure
Research Journal of Applied Sciences, Engineering and Technology 6(1): 160-164, 2013 ISSN: 2040-7459; e-issn: 2040-7467 Maxwell Scientific Organization, 2013 Subitted: Noveber 08, 2012 Accepted: Deceber
More informationWeak and Semi-strong Form Stock Return Predictability Revisited
Weak and Sei-strong For Stock Return Predictability Revisited WAYNE E. FERSON ANDREA HEUSON TIE SU Boston College 140 Coonwealth Avenue, Chestnut Hill, MA. 02467 University of Miai 5250 University Drive,
More informationSome estimates for income elasticities of leisure activities in the United States
MPRA Munich Personal RePEc Archive Soe estiates for incoe elasticities of leisure activities in the United States Jorge González Chapela Centro Universitario de la Defensa de Zaragoza 14. July 2014 Online
More informationNo 70. Male-female labour market participation and wage differentials in Greece
CENTRE OF PLANNING AND ECONOMIC RESEARCH No 70 Male-feale labour arket participation and wage differentials in Greece C. N. Kanellopoulos K. G. Mavroaras February 1999 C. N. Kanellopoulos Senior Research
More informationAn Index Number Formula Problem: the Aggregation of Broadly Comparable Items
An Index Nuber Forula Proble: the Aggregation of Broadly Coparable Ites Mick Silver* International Monetary Fund Presentation to the (Ottawa) International Working Group on Price Indices (May 27 29, 2009)
More informationNBER WORKING PAPER SERIES THE LEVERAGE EFFECT PUZZLE: DISENTANGLING SOURCES OF BIAS AT HIGH FREQUENCY. Yacine Ait-Sahalia Jianqing Fan Yingying Li
NBER WORKING PAPER SERIES THE LEVERAGE EFFECT PUZZLE: DISENTANGLING SOURCES OF BIAS AT HIGH FREQUENCY Yacine Ait-Sahalia Jianqing Fan Yingying Li Working Paper 17592 http://www.nber.org/papers/w17592 NATIONAL
More informationLast For A Lifetime. Making Your Money. Why You Need to Know About Annuities
Making Your Money Last For A Lifetie Why You Need to Know About Annuities A Joint Project of The Actuarial Foundation and WISER, the Woen s Institute for a Secure Retireent Acknowledgeents Special thanks
More informationComparison of OLS and LAD regression techniques for estimating beta
Comparison of OLS and LAD regression techniques for estimating beta 26 June 2013 Contents 1. Preparation of this report... 1 2. Executive summary... 2 3. Issue and evaluation approach... 4 4. Data... 6
More informationProvided in Cooperation with: Center for Financial Studies (CFS), Goethe University Frankfurt
econstor www.econstor.eu Der Open-Access-Publikationsserver der ZBW Leibniz-Inforationszentru Wirtschaft The Open Access Publication Server of the ZBW Leibniz Inforation Centre for Econoics Schidt, Daniel
More informationLECTURE 4: MIXED STRATEGIES (CONT D), BIMATRIX GAMES
LECTURE 4: MIXED STRATEGIES (CONT D), BIMATRIX GAMES Mixed Strategies in Matrix Gaes (revision) 2 ixed strategy: the player decides about the probabilities of the alternative strategies (su of the probabilities
More informationProduction, Process Investment and the Survival of Debt Financed Startup Firms
Babson College Digital Knowledge at Babson Babson Faculty Research Fund Working Papers Babson Faculty Research Fund 00 Production, Process Investent and the Survival of Debt Financed Startup Firs S. Sinan
More informationVolume Title: Bank Stock Prices and the Bank Capital Problem. Volume URL:
This PDF is a selection from an out-of-print volume from the National Bureau of Economic Research Volume Title: Bank Stock Prices and the Bank Capital Problem Volume Author/Editor: David Durand Volume
More informationLearning Objectives CMT Level III
Learning Objectives CMT Level III - 2018 The Integration of Technical Analysis Section I: Risk Management Chapter 1 System Design and Testing Explain the importance of using a system for trading or investing
More informationEconomic Growth, Inflation and Wage Growth: Experience from a Developing Country
www.sciedu.ca/br Business and Manageent Research Vol., No. ; 0 Econoic Growth, Inflation and Wage Growth: Experience fro a Developing Countr Shahra Fattahi (Corresponding author) Departent of Econoics
More informationm-string Prediction
Figure 1. An =3 strategy. -string Prediction 000 0 001 1 010 1 011 0 100 0 101 1 110 0 111 1 10 Figure 2: N=101 s=1 9 8 7 6 σ 5 4 3 2 1 0 0 2 4 6 8 10 12 14 16 42 Figure 3: N=101 s=2 15 10 σ 5 0 0 2 4
More informationOccasional Paper. Risk Measurement Illiquidity Distortions. Jiaqi Chen and Michael L. Tindall
DALLASFED Occasional Paper Risk Measurement Illiquidity Distortions Jiaqi Chen and Michael L. Tindall Federal Reserve Bank of Dallas Financial Industry Studies Department Occasional Paper 12-2 December
More informationARTICLE IN PRESS. Pricing in debit and credit card schemes. Julian Wright* 1. Introduction
ARTICLE IN PRE Econoics Letters x (200) xxx xxx www.elsevier.co/ locate/ econbase Pricing in debit and credit card schees Julian Wright* Departent of Econoics, University of Auckland, Private ag 92019,
More informationIntroduction to Statistical Inference
Itroductio to Statistical Iferece Fial Review CH1: Picturig Distributios With Graphs 1. Types of Variable -Categorical -Quatitative 2. Represetatios of Distributios (a) Categorical -Pie Chart -Bar Graph
More informationUnisex-Calculation and Secondary Premium Differentiation in Private Health Insurance. Oliver Riedel
Unisex-Calculation and Secondary Preiu Differentiation in Private Health Insurance Oliver Riedel University of Giessen Risk Manageent & Insurance Licher Strasse 74, D - 35394 Giessen, Gerany Eail: oliver.t.riedel@wirtschaft.uni-giessen.de
More informationProject selection by using AHP and Bernardo Techniques
International Journal of Huanities and Applied Sciences (IJHAS) Vol. 5, No., 06 ISSN 77 4386 Project selection by using AHP and Bernardo Techniques Aza Keshavarz Haddadha, Ali Naazian, Siaak Haji Yakhchali
More informationCONTENTS Advisory Setup and Installation Wired Hand Control year Warranty Manufactured by I nnova S leep Systems 1-14
PB270 Wired Adjustable Base Owner Manual CONTENTS Advisory Setup and Installation Wired Hand Control 1-2-20 year Warranty Manufactured by I nnova S leep Systes 1-14 WARNING Attention: Read the following
More informationCorrective Taxation versus Liability
Aerican Econoic Review: Papers & Proceedings 2011, 101:3, 273 276 http://www.aeaweb.org/articles.php?doi=10.1257/aer.101.3.273 Law and Econoics Corrective Taxation versus Liability By Steven Shavell* Since
More informationExempt Organization Business Income Tax Return (and proxy tax under section 6033(e))
6/7/216 1:39:49 AM 1 214 Return COMBINED JEWISH PHILANTHROPIES OF For 99-T PUBLIC DISCLOSURE COPY Exept Organization Business Incoe Tax Return (and proxy tax under section 633(e)) OMB No. 1545-687 214
More informationMarket Power and Policy Effects in the Korean Infant Formula Industry - Case Study-
Market ower and olicy Effects in the Korean Infant Forula Industry - Case Study- 1. Infant forula industry in Korea Over 90% of new-born babies are priarily fed infant forula There are only four copanies
More informationQED. Queen s Economics Department Working Paper No. 1088
QED Queen s Econoics Departent Working Paper No. 1088 Regulation and Taxation of Casinos under State-Monopoly, Private Monopoly and Casino Association Regies Hasret Benar Eastern Mediterranean University
More informationFINITE SAMPLE DISTRIBUTIONS OF RISK-RETURN RATIOS
Available Online at ESci Journals Journal of Business and Finance ISSN: 305-185 (Online), 308-7714 (Print) http://www.escijournals.net/jbf FINITE SAMPLE DISTRIBUTIONS OF RISK-RETURN RATIOS Reza Habibi*
More informationState of Delaware VOYA PLAN and Your Voya Retirement Insurance and Annuity Company Investment Program - Plan-related Information
State of Delaware VOYA PLAN 664093 and 664094 Your Voya Retireent Insurance and Annuity Copany Investent Progra - Plan-related Inforation August 17,2016 The purpose of this docuent is to suarize certain
More informationMETHODOLOGY AND RESULTS
Methodological Docuent METHODOLOGY AND RESULTS PRODUCER PRICE INDEX IN BOSNIA AND HERZEGOVINA Bosnia and Herzegovina Agenc for Statistics of Bosnia and Herzegovina Federal Office of Statistics of Federation
More informationAggregation Issues in Operational Risk
Aggregation Issues in Operational Ris Rosella Giacoetti Departent of Matheatics, Statistics, Coputer Science and Applications L. Mascheroni School of Econoics and Business Adinistration, University of
More informationMarket Timing Ability and Stock Selection Skills of the Fund Manager
CHAPTER 6 Market Timing Ability and Stock Selection Skills of the Fund Manager Chapter 6 Market Timing Ability of the Fund Manager Page 148 MARKET TIMING ABILITY AND STOCK SELECTION SKILLS 6.1 Introduction
More informationA NOJE;o.n INTEREST RATE RISK, SYSTEMATIC RISK and the PLANNING. Researchmeraorandum Sept. '85
ET 05548 1985 023 SERIE RE5ERR[HE0RHHDH A NOJE;o.n INTEREST RATE RISK, SYSTEMATIC RISK and the PLANNING HORIZON Leon J. de Man Researcheraorandu 1985-23 Sept. '85 VRIJE UNIVERSITEIT Faculteit der Econoische
More informationHedge Fund Volatility: It s Not What You Think It Is 1 By Clifford De Souza, Ph.D., and Suleyman Gokcan 2, Ph.D. Citigroup Alternative Investments
Disclaimer: This article appeared in the AIMA Journal (Sept 2004), which is published by The Alternative Investment 1 Hedge Fd Volatility: It s Not What You Think It Is 1 By Clifford De Souza, Ph.D., and
More information