Review of Business and Economics Studies, 2018, том 6, № 1
Покупка
Основная коллекция
Тематика:
Экономика. Бухгалтерский учет. Финансы
Наименование: Review of Business and Economics Studies
Год издания: 2018
Кол-во страниц: 92
Дополнительно
Вид издания:
Журнал
Артикул: 705326.0001.99
Тематика:
ББК:
УДК:
ГРНТИ:
Скопировать запись
Фрагмент текстового слоя документа размещен для индексирующих роботов.
Для полноценной работы с документом, пожалуйста, перейдите в
ридер.
Review of Business and Economics Studies EDITOR-IN-CHIEF Prof. Alexander Ilyinsky Dean, International Finance Faculty, Financial University, Moscow, Russia ailyinsky@fa.ru EXECUTIVE EDITOR Dr. Zbigniew Mierzwa EDITORIAL BOARD Dr. Mark Aleksanyan Adam Smith Business School, The Business School, University of Glasgow, UK Prof. Edoardo Croci Research Director, IEFE Centre for Research on Energy and Environmental Economics and Policy, Università Bocconi, Italy Prof. Moorad Choudhry Dept.of Mathematical Sciences, Brunel University, UK Prof. David G. Dickinson Department of Economics, Birmingham Business School, University of Birmingham, UK Prof. Chien-Te Fan Institute of Law for Science and Technology, National Tsing Hua University, Taiwan Prof. Wing M. Fok Director, Asia Business Studies, College of Business, Loyola University New Orleans, USA Prof. Konstantin P. Glushchenko Faculty of Economics, Novosibirsk State University, Russia Prof. George E. Halkos Associate Editor in Environment and Development Economics, Cambridge University Press; Director of Operations Research Laboratory, University of Thessaly, Greece Dr. Christopher A. Hartwell President, CASE — Center for Social and Economic Research, Warsaw, Poland Prof. Sebastian Jaimungal Associate Chair of Graduate Studies, Dept. Statistical Sciences & Mathematical Finance Program, University of Toronto, Canada Prof. Vladimir Kvint Chair of Financial Strategy, Moscow School of Economics, Moscow State University, Russia Prof. Alexander Melnikov Department of Mathematical and Statistical Sciences, University of Alberta, Canada Prof. George Kleiner Deputy Director, Central Economics and Mathematics Institute, Russian Academy of Sciences, Russia Prof. Kern K. Kwong Director, Asian Pacific Business Institute, California State University, Los Angeles, USA Prof. Dimitrios Mavrakis Director, Energy Policy and Development Centre, National and Kapodistrian University of Athens, Greece Prof. Stephen McGuire Director, Entrepreneurship Institute, California State University, Los Angeles, USA Prof. Rustem Nureev Сhairman for Research of the Department of Economic Theory, Financial University, Russia Dr. Oleg V. Pavlov Associate Professor of Economics and System Dynamics, Department of Social Science and Policy Studies, Worcester Polytechnic Institute, USA Prof. Boris Porfiriev Deputy Director, Institute of Economic Forecasting, Russian Academy of Sciences, Russia Prof. Thomas Renstrom Durham University Business School, Department of Economics and Finance, Durham University Prof. Alan Sangster Professor of Accounting (Business and Management) at University of Sussex, UK Prof. Svetlozar T. Rachev Professor of Finance, College of Business, Stony Brook University, USA Prof. Boris Rubtsov Deputy chairman of Department of financial markets and banks for R&D, Financial University, Russia Dr. Shen Minghao Director of Center for Cantonese Merchants Research, Guangdong University of Foreign Studies, China Prof. Dmitry Sorokin Chairman for Research, Financial University, Russia Prof. Robert L. Tang Chancellor for Academic, De La Salle College of Saint Benilde, Manila, The Philippines Dr. Dimitrios Tsomocos Saïd Business School, Fellow in Management, University of Oxford; Senior Research Associate, Financial Markets Group, London School of Economics, UK REVIEW OF BUSINESS AND ECONOMICS STUDIES (ROBES) is the quarterly peerreviewed scholarly journal published by the Financial University under the Government of Russian Federation, Moscow. Journal’s mission is to provide scientific perspective on wide range of topical economic and business subjects. CONTACT INFORMATION Financial University Leningradsky prospekt, 53, office 5.6 123995 Moscow Russian Federation Telephone: +7 (499) 943-98-02 Website: www.robes.fa.ru AUTHOR INQUIRIES Inquiries relating to the submission of articles can be sent by electronic mail to robes@fa.ru. COPYRIGHT AND PHOTOCOPYING © 2017 Review of Business and Economics Studies. All rights reserved. No part of this publication may be reproduced, stored or transmitted in any form or by any means without the prior permission in writing from the copyright holder. Single photocopies of articles may be made for personal use as allowed by national copyright laws. ISSN 2308-944X
Вестник исследований бизнеса и экономики ГЛАВНЫЙ РЕДАКТОР А.И. Ильинский, профессор, декан Международного финансо вого факультета Финансового университета ВЫПУСКАЮЩИЙ РЕДАКТОР Збигнев Межва, д-р экон. наук РЕДАКЦИОННЫЙ СОВЕТ М.М. Алексанян, профессор Бизнесшколы им. Адама Смита, Университет Глазго (Великобритания) К. Вонг, профессор, директор Института азиатско-тихоокеанского бизнеса Университета штата Калифорния, Лос-Анджелес (США) К.П. Глущенко, профессор экономического факультета Новосибирского госуниверситета С. Джеимангал, профессор Департамента статистики и математических финансов Университета Торонто (Канада) Д. Дикинсон, профессор Департамента экономики Бирмингемской бизнесшколы, Бирмингемский университет (Великобритания) В.Л. Квинт, заведующий кафедрой финансовой стратегии Московской школы экономики МГУ, профессор Школы бизнеса Лассальского университета (США) Г. Б. Клейнер, профессор, член-корреспондент РАН, заместитель директора Центрального экономико-математического института РАН Э. Крочи, профессор, директор по научной работе Центра исследований в области энергетики и экономики окружающей среды Университета Боккони (Италия) Д. Мавракис, профессор, директор Центра политики и развития энергетики Национального университета Афин (Греция) С. Макгвайр, профессор, директор Института предпринимательства Университета штата Калифорния, Лос-Анджелес (США) А. Мельников, профессор Депар та мента математических и ста тистических исследований Университета провинции Альберта (Канада) Р.М. Нуреев, профессор, научный руководитель Департамента экономической теории Финансового университета О.В. Павлов, профессор Депар та мента по литологии и полити ческих исследований Ворчестерского политехнического института (США) Б.Н. Порфирьев, профессор, член-корреспондент РАН, заместитель директора Института народнохозяйственного прогнозирования РАН С. Рачев, профессор Бизнес-кол леджа Университета Стони Брук (США) Т. Ренстром, профессор, Школа Бизнеса Даремского университета, Департамент Экономики и Финансов Б.Б. Рубцов, профессор, заместитель руководителя Департамента финансовых рынков и банков по НИР Финансового университета А. Сангстер, профессор, Сассекский университет (Великобритания) Д.Е. Сорокин, профессор, членкорреспондент РАН, научный руководитель Финансового университета Р. Тан, профессор, ректор Колледжа Де Ла Саль Св. Бенильды (Филиппины) Д. Тсомокос, Оксфордский университет, старший научный сотрудник Лондонской школы экономики (Великобритания) Ч.Т. Фан, профессор, Институт права в области науки и технологии, национальный университет Цин Хуа (Тайвань) В. Фок, профессор, директор по исследованиям азиатского бизнеса Бизнес-колледжа Университета Лойола (США) Д.Е. Халкос, профессор, Университет Фессалии (Греция) К.А. Хартвелл, президент Центра социальных и экономических исследований CASE (Польша) М. Чудри, профессор, Университет Брунеля (Великобритания) М. Шен, декан Центра кантонских рыночных исследований Гуандунского университета (КНР) Редакция научных журналов Финансового университета 123995, Москва, ГСП-5, Ленинградский пр-т, 53, комн. 5.6 Тел. 8 (499) 943-98-02. Интернет: www.robes.fa.ru. Журнал “Review of Business and Economics Studies” («Вест ник исследований бизнеса и экономики») зарегистрирован в Федеральной службе по надзору в сфере связи, информационных технологий и массовых коммуникаций 15 сентября 2016 г. Свидетельство о регистрации ПИ № ФС77-67072. Подписано в печать: 14.03.2018. Формат 60 × 84 1/8. Заказ № 238 от 14.03.2018. Отпечатано в Отделе полиграфии Финуниверситета (Ленинградский проспект, д. 49). 16+
CONTENTS Optimization of the Shape of the Pareto Set in the Problems of Multi-criterial Programming Bula Katendi Axel, E. A. Umnov, A. E. Umnov . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .5 Does Enterprise Value Really Depend on WACC and Free Cash Flow? The Evidence of Irrationality from the Oil and Gas Sector Pavel E. Zhukov . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .17 Weather Derivatives in Russia: Farmers’ Insurance against Temperature Fluctuations Eric Carkin, Stanislav Chekirov, Anastasia Echimova, Caroline Johnston, Congshan Li, Vladislav Secrieru, Alyona Strelnikova, Marshall Trier, Vladislav Trubnikov . . . . . . . .29 Does Social Inequality Stimulate the Economic Growth? (On the examples of the chosen developing countries) Alina Pukhaeva, Elena Miroshina (Silantieva) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .43 Country Risk in International Investment Its’ Structure and Methods of Estimation Alexey Ivkin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .56 Business Valuation of Nike Elena Miroshina (Silantieva), Egor Romanov . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .78 “Uniform Subsidy” and New Trends in Financing of Agricultural Insurance in Russian Federation L.Yu. Piterskaya, N. A. Tlisheva, A. V. Piterskaya . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .83 Review of Business and Economics Studies Volume 6, Number 1, 2018
Вестник исследований бизнеса и экономики № 1, 2018 CОДЕРЖАНИЕ Оптимизация формы множества Парето в задачах многокритериального программирования Була Аксел Катенди, Е. А. Умнов, А. Е. Умнов . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 Разве стоимость компании действительно зависит от средневзвешенной стоимости капитала и свободного денежного потока? Свидетельства иррациональности в нефтегазовом секторе Павел Е. Жуков . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17 Погодные деривативы в России: страхование фермеров от колебаний температуры Eric Carkin, Станислав Чекиров, Анастасия Екимова, Caroline Johnston, Congshan Li, Vladislav Secrieru, Алена Стрельникова, Marshall Trier, Владислав Трубников . . . . . 29 Стимулирует ли социальное неравенство экономический рост? (на примерах выбранных развивающихся стран) Алина Пухаева, Елена Мирошина (Силантьева) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43 Страновой риск в международных инвестициях: структура и методы расчета Алексей Ивкин . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56 Оценка бизнеса корпорации Nike Елена Мирошина (Силантьева), Егор Романов . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78 «Единая субсидия» и новые веяния в субсидировании агрострахования в Российской Федерации Людмила Питерская, Нафсэт Тлишева, Анастасия Питерская . . . . . . . . . . . . . . . . 83
Review of Business and Economics Studies doi: 10.26794/2308-944X-2018-6–1-5–16 2018, Vol. 6, No. 1, 5–16 Optimization of the Shape of the Pareto Set in the Problems of Multi-criterial Programming Bula Katendi Axel1, E. A. Umnov2, A. E. Umnov2 Moscow institute of Physics and Technology (State University) 1axelbula@gmail.com, 2mail@umnov.ru Abstract in this paper, a scheme for using the method of smooth penalty functions for the dependence of solutions of multi-criterial optimization problems on parameters is being considered. in particular, algorithms based on the method of smooth penalty functions are given to solve problems of optimization by the parameters of the level of consistency of the objective functions and to find the corresponding shape of the Pareto’s set. Keywords: multi-criterial parametrical programming tasks; set of Pareto; method of smooth penalty functions; optimization problem in terms of parameters. JEL classification: C61, C65 I n mathematical modeling, it is often necessary to formalize preferences for states of the modeled object that generates several independent target functions. According to the historically established tradition, in this case, it is customary to talk about of multi-criterial optimization problems. A finite-dimensional multi-criterial model is a mathematical model with N objective functions: ( ) [ ] , max1, k x f x u k N → = , (1) subject to maximization possessing at interior points of the set of elements n x E ∈ , and satisfying the following conditions: ( ) [ ] , 01, iy x u i m ≤ = (2) where r u E ∈Θ ⊆ — vector of parameters of the model. It is assumed that the functions ( ) , kf x u and ( ) , iy x u are sufficiently smooth, i. e. they have continuous derivatives of a desired order in all their arguments. The incorrectness in the general case of such a statement is obvious, since the element x that is extremal for one of objective functions, in general, is not such for others. However, useful information can be obtained by successively solving the following problems with a criterion for finding an extremum on the set (2) of each of the functions (1) separately for [ ] 1, k N = : ( ) , max k x f x u → subject to ( ) [ ] , 01, iy x u i m ≤ = (3) The objective function ( ) , kf x u is called improvable in the feasible point 0 x (i. e., satisfying the condition (3)) if there is another feasible point 1x , for which ( ) ( ) 1 0 , , k k f x u f x u > . It is clear that the solution of problem (3) for any k = [1, N] is un-improvable, or “non-ideal” at the point of view of the other objective function ( ) , kf x u .
The concept of improving the multi-criterial objective function allows the feasible points to be divided into two subsets: for the first, all feasible points improve all objective functions and for the second, there are points for which the improvement of one function causes the deterioration of at least one other function. The second subset is called a Pareto-type set or, simply, a Pareto set. A general universal approach to the solution of multi-criterial optimization problems has not been proposed yet, but numerous approaches have been developed (see Fiacco & McCormick, 1968; Lotov & Pospelov, 2008), which limit the number of solutions. For example, in the practical use of multi-criterial mathematical models, the set of independent objective functions is often replaced by a single one, thus passing to the standard problem of mathematical programming, allowing finding consistent or compromising solutions on the Pareto set in a certain sense. Statement of the problem In this article, the problem of finding an element on the set (2) that minimizes the gap between the objective functions will be considered as a compromise. In other words, this is a mathematical programming problem of the following form: minimize ρ, subject to ρ≥ 0, ( ) [ ] , 01, , iy x u i m ≤ = (4) ( ) ( ) [ ] * , 1,k k f x u F u k N ≥ −ρ = whose solution will be denoted by ( ) ** u ρ and ( ) ** x u . Here ( ) ( ) ( ) * * , k k k F u f x u u = and ρ is called “mismatch value”. The problem (4) is naturally called a two-level parametric problem since in its formulation it contains solutions ( ) [ ] * 1,k F u k N = of problems with single criterion (3), which we call first level problems. In this case, both in the problems of the first and the second level, it is assumed that the vector of the parameters u∈Θ is fixed. It is clear that the extreme value of the mismatch between the criteria in the general case is determined by the properties of the Pareto set and depends on the parameters vector u . Therefore, it is natural to indicate the third level optimization problem for the models (1)–(2) as follows: optimize the expression ( ) ** u ρ subject to u∈Θ (5) which solution will be the vector of parameters *** u ∈Θ and the number ( ) *** ** *** . u ρ = ρ In the present paper, possible solutions to problem (5) will be considered. Solution method Let us consider the problem of finding in the parameter space a standard method (for example, gradient) of finding the extremum of the mismatch value of the objective functions of the multicriterial model (3)–(4)–(5). The specificity of this problem is based on the fact that the formulation of the problem (5) (the upper level or third level) includes the dependence ( ) ** u ρ , the solution of the problem (4) (the second level) which in turn, depends on ( ) ( ) ( ) [ ] * * , 1,k k k F u f x u u k N = ∀ = — the solutions of the problem (3) (the lower level or first level). The functions ( ) ** u ρ and ( ) [ ] * 1,k F u k N ∀ = in the general case (even for smooth functions ( ) , kf x u and ( ) , iy x u ) may be not differentiable, that why the use of any numerical method based on Taylor approximations is not possible. Optimization of the Shape of the Pareto Set in the Problems of Multi-criterial Programming
It is proposed the use of the method of smooth penalty function to overcome this difficulty (see Umnov, 1975) and obtain a sufficiently smooth approximation dependences of ( ) ** u ρ and ( ) [ ] * 1,k F u k N ∀ = . It’s assumed that the penalty function ( ) , P s τ , which penalizes the restriction 0 s ≤ , satisfies the following conditions: 1 0 ∀τ > and s ∀ , the function ( ) , P s τ has continuous derivatives with respect to all its arguments up to the second order 2 0 ∀τ > and s ∀ , 2 2 0;0P P s s ∂ ∂ > > ∂ ∂ (6) 3 ( ) , 0P s s τ > ∀ and 0 ∀τ > , and, ( ) 0 ,0 lim , 0,0 s P s s τ→+ +∞ > τ = < (7) When solving the third-level problem by an iterative method, for each step of the method, it is necessary preliminary to solve the problems of the second and first levels for a fixed vector of parameters u . Let us first consider a possible scheme for solving first-level problems. In fact, we will use an auxiliary function for the one-criterion problems (3), as follows: ( ) ( ) ( ) ( ) [ ] 1 , , , , , 1,m k k i i A x u f x u P y x u k N = τ = − τ ∀ ∈ ∑ (8) while a sufficiently smooth penalty function ( ) , P s τ satisfies conditions (6) and (7). As shown in Zhadan (2014), instead of the smooth approximations ( ) * k x u solutions of each task of problem (3), we can take ( ) k x u stationary points of the auxiliary function (8), defined like: [ ] ( ) 01,9 k j A j n x ∂ = ∀ ∈ ∂ (9) or [ ] 1 01,m k i j i j i f y P j n x y x = ∂ ∂ ∂ − = ∀ ∈ ∂ ∂ ∂ ∑ Since the condition of the second-level problem (4) includes the dependencies ( ) ( ) ( ) [ ] * * , 1,k k k F u f x u u k N = ∀ = which are not differentiable functions for all their arguments, then for these dependencies it is also necessary to choose a smoothed approximation. As an approximation, the auxiliary function calculated at a stationary point ( ) ( ) ( ) , , k k k F u A x u u = τ can be used, because (due to the properties of the penalty function method) its value for small positive τ is close to the optimal value of the objective function of the k -th problem (3). Standard optimization methods used for lower-level tasks, based on the use of continuous gradients or other differential characteristics, suggest that in addition to the solving system (9), these characteristics themselves can be found. Let us demonstrate this using the example of calculating the derivatives of the function ( ) k F u with respect to the components of the vector u of parameters. As ( ) ( ) ( ) , , k k k F u A x u u = τ , then according to the rule for differentiating a composite functionof several variables, we have: Optimization of the Shape of the Pareto Set in the Problems of Multi-criterial Programming
[ ] 1 1,n j k k k p p j p j x F A A p r u u x u = ∂ ∂ ∂ ∂ = + ∀ ∈ ∂ ∂ ∂ ∂ ∑ Using (9), we have: ( ) ( ) [ ] , , 1,k k k p p F A x u u p r u u ∂ ∂ = τ ∀ ∈ ∂ ∂ (10) Note that the last simplification would be impossible if for ( ) * k F u a more natural approximation ( ) ( ) * , k k f x u u is used instead of the smoothing approximation ( ) ( ) , , k k A x u u τ . Let us now look into the solution to the second-level problem. To make application of the penalty function method more convenient, the problem (4) is expressed as: maximize −ρ , subject to −ρ ≥ 0, ( ) [ ] , 01, , iy x u i m ≤ = (11) ( ) [ ] , , 01,k Y x u k N ρ ≤ = where ( ) ( ) ( ) * , , , k k k Y x u F u f x u ρ = −ρ− The solution to this problem will be denoted by ( ) ** u ρ and ( ) ** x u . Let us define the auxiliary function for the problem (10) as follows: ( ) ( ) ( ) ( ) ( ) ( ) 1 1 , , , , , , , , , N m k i k i E x u P P Y x u P y x u = = τ ρ = −ρ− τ −ρ − τ ρ − τ ∑ ∑ (12) replacing previously in ( ) , , k Y x u ρ the dependency ( ) * k F u by its smoothed approximation ( ) k F u . For the set of variables { } 1 2 ,,,,n x x x ρ … , the conditions for the stationarity of the auxiliary function (12) will be: [ ] 1 1 1 1 0 0 1, = = = ∂ ∂ ∂ = − + + = ∂ ∂ ∂ ∂ ∂ ∂ ∂ ∂ = − = ∀ = ∂ ∂ ∂ ∂ ∂ ∑ ∑ ∑ N k k N m k i k i j k j i j E P P Y f y E P P j N x Y x y x ρ ρ (13) Let the solutions of system (13) be ( ) u ρ and ( ) x u , then, as a smoothed approximation of the dependency ( ) ** u ρ , we can use the function ( ) ( ) ( ) ( ) , , , E u E u x u u = − ρ ρ . The derivatives of this function by the components of the vector u ofparameters and the rule for differentiating a composite function of several variables give us: [ ] 1 1,n j p p j p p j x E E E E p r u u x u u = ∂ ∂ ∂ ∂ ∂ ∂ρ = + + ∀ ∈ ∂ ∂ ∂ ∂ ∂ρ ∂ ∑ . From (13) we know that 0 E ∂ = ∂ρ and [ ] 01,j E j N x ∂ = ∀ = ∂ . Then the last expression can be written simply: Optimization of the Shape of the Pareto Set in the Problems of Multi-criterial Programming
( ) ( ) ( ) [ ] , , , 1,p p E E u x u u p r u u ∂ ∂ = ρ ρ ∀ ∈ ∂ ∂ (14) Finally, we obtain formulas for the gradient components of ( ) E u in terms of the functions usedin the formulation of the multicriterial model (4)–(5) and the method of smooth penalty functions. From (12) it is obtained: 1 1 N N k i p k p i p k k Y y E P P u Y u y u = = ∂ ∂ ∂ ∂ ∂ = − − ∂ ∂ ∂ ∂ ∂ ∑ ∑ where k k k p p p Y F f u u u ∂ ∂ ∂ = − ∂ ∂ ∂ and the value of k p F u ∂ ∂ can be found from (10). Formulas (14) allow us to solve the third-level problem by applying any of the first-order methods, for example, conjugate directions. Note that second-order methods should also be considered here. However, this will be done at the end of the article, while now let us illustrate an example. Proposed method in use Let us consider multi-criterial mathematical model in which 3 1 2 3 T x x x x E = ∈ is a vector of independent variables and 2 1 2 T u u u E = ∈ is a vector of parameters. The problem is to maximize for x and u∈Θ the functions: ( ) ( ) ( ) 1 1 2 2 2 3 , ,, ,, f x u x f x u x f x u x = = = subject to ( ) ( ) ( ) ( ) 1 2 3 1 1, 2 1 2 1, 2 2 3 1, 2 3 1, 2 0,0,0,x x x a u u x a u u x a u u x b u u ≥ ≥ ≥ + + ≤ and where the functions ( ) ( ) ( ) ( ) 1 1, 2 2 1, 2 3 1, 2 1, 2 ,,a u u a u u a u u and b u u are given by the condition below. A valid region of the model (with an allowable fixed u ) is a rectangular pyramid OABC. The Pareto set coincides with the face of ABC or is a part of it. 100 T A u = and 2 00 T B u = . We assume that the set Θ in the parameter space is given by the condition that the sum of the lengths of the segments OA, OB and OC is constant and equal 3. Figure 1. Geometric interpretation of models (6)–(7). Optimization of the Shape of the Pareto Set in the Problems of Multi-criterial Programming
Applying the standard methods of analytic geometry, we find that for the compatibility of the system of model constraints, the existence of 0 r ≥ is necessary such that: ( ) 1 1 2 2 , a u u u r = ( ) 2 1 2 1 , a u u u r = ( ) 3 1 2 1 2 , a u u u u = ( ) 1 2 1 2 , b u u u u r = By choosing r such that 1 2 3 r u u = − − , we assure that the set Θ will not be empty when 1 0.1 2.5 u ≤ ≤ and 2 0.1 2.5 u ≤ ≤ . The minimum value of the discrepancy between the criteria in this example depends on the form of the Pareto set, which is the triangle ABC, or a part of it. A graphic representation of the dependence of the error value of the objective functions on the parameters 1u and 2u is shown in pictures 2 and 3. Let us see with more details the properties of this dependence. Clearly, the solutions of the first-level problems (3) for fixed 1u and 2 u are: ( ) ( ) ( ) ( ) ( ) ( ) * * * 1 1 2 2 3 1 2 ,,3 . f x u u f x u u f x u u u = = = − − Consequently, the task of the second level (4) — minimizing the discrepancy of the criteria, will have the form: minimize ρ according to { } 1 2 3 , ,, x x x ρ subject to 0 ρ ≥ , 1 2 3 0,0,0,x x x ≥ ≥ ≥ 2 1 1 2 1 2 3 1 2 , u rx u rx u u x u u r + + ≤ 1 1 , x u ≥ −ρ 2 2 , x u ≥ −ρ 3x r ≥ −ρ , 1 2 3 r u u = − − His solution will be designed by ( ) ** 1 2 , . u u ρ Finally, the task of the third level (5) for our example will be: minimize ( ) ** 1 2 , u u ρ by { } 1 2 , u u when 1 0.1 2.5 u ≤ ≤ and 2 0.1 2.5 u ≤ ≤ . It is known from the theory of mathematical programming that the properties of the dependence ( ) ** 1 2 , u u ρ are primarily determined by how the set of constraints of a model of the «inequality» type is divided into active and inactive ones, that is, the first of which are satisfied as equalities, and the second — as strict inequalities. This separation depends on the values of the parameters of the model and its optimal variant determines the solution of the second-level problem. First, suppose that the values of the model parameters initiate a conflict of all three criteria simultaneously. In other words, the improvement of the value of any one of the objective functions of the model is possible only if the values of all the others deteriorate. In this case, the last five constraints of the second-level problem must be active, and we obtain Optimization of the Shape of the Pareto Set in the Problems of Multi-criterial Programming
the following system of equations, which allows us to find the analytical form of the dependence ( ) ** 1 2 , u u ρ . 2 1 1 2 1 2 3 1 2 1 1 2 2 3 1 2 3 u rx u rx u u x u u r x u x u x r r u u + + = = −ρ = −ρ = −ρ = − − Solving this system above, we have the analytic form of ρwhich depends on 1u and 2 u : ( ) ** 1 2 1 2 1 2 2 , 1 1 1 3 u u u u u u ρ = + + − − The stationary points of ( ) ** 1 2 , u u ρ are 11,33 ,33 T T T − − and 33 T can be easily found, and for the first point the function has a local maximum with the value 2/3 according to the Sylvester criterion, not for the others because they don’t satisfy the condition of non-negativity of the variables 1 2 3 ,,x x x and r . The formula obtained is valid only in a certain area contained in Θ . An analysis of the isoline system shown in Picture 3, allows selecting five areas with different sets of active restrictions. Light lines determine the boundaries between the areas. The formula obtained above is valid only in area 4. In this area, the Pareto set of the model under consideration consists of the interior points of the triangle ABC. Outside area 4, the formula for ( ) ** 1 2 , u u ρ is different. For the area 1, for example, ( ) ** 1 2 , u u ρ is found from the system of equations: 2 1 1 2 1 2 3 1 2 1 2 2 3 1 2 03 u rx u rx u u x u u r x x u x r r u u + + = = = −ρ = −ρ = − − Figure 2. The quantity OA OB OC + + is constant (graphic — 3D). Optimization of the Shape of the Pareto Set in the Problems of Multi-criterial Programming
since the set of active restrictions in it is different: it contains the condition 1 0 x = , instead of 1 1 x u = −ρ . We can easily prove that in the area 1 ( ) ** 1 2 2 1 2 2 , 1 1 3 u u u u u ρ = + − − There are not stationary points for this dependency. For areas 2 and 3, the arguments and results are similar. The Pareto sets in areas 1, 2, and 3 are the sides of the triangle ABC: BC, AC, and AB, respectively. Finally, we note that in area 5 the system of conditions (2) is contradictory. In this case, the exact solution to the problem of the upper (third) level has the form: ** ** ** 1 2 2 1,1,3 u u = = ρ = Let us now describe the method for solving the third-level problem for the variant of the multicriterial model. We have: objective functions: maximize by 3 1 2 3 T x x x x E = ∈ ( ) 1 1 , f x u x = ( ) 2 2 , f x u x = ( ) 3 3 , f x u x = subject to 1 2 3 0,0,0, x x x ≥ ≥ ≥ ( ) ( ) ( ) ( ) 1 1 2 2 3 3 a u x a u x a u x b u + + ≤ where ( ) ( ) ( ) ( ) ( ) ( ) ( ) 1 2 1 2 2 1 1 2 3 1 2 1 2 1 2 3 3 3 a u u u u a u u u u a u u u b u u u u u = − − = − − = = − − Figure 3. The quantity OA OB OC + + is constant (isoline view). Optimization of the Shape of the Pareto Set in the Problems of Multi-criterial Programming