22
経済の数理解析 京都大学数理解析研究所 研究集会 京都大学数理解析研究所の共同研究事業の一つとして、下記のように研究集会を催しますので、 ご案内申し上げます。 研究代表者 丸山 徹(慶應義塾大学) 平成17年2月11日(金)-2月13日(日) 京大会館 210号室 〒606-8305 京都市左京区吉田河原町15-9 ℡075-751-8311 市バス 京大正門前 下車/京都バス 荒神橋 下車 京阪電車鴨東線丸太町駅下車徒歩約10分 講演時間 60分(うち質疑応答10分) プログラム 2月11日(金) * 印:講演者 午前の部 座長 : 丸山 徹(慶應義塾大学) 9:00-10:00 加藤 寛之(慶應義塾大学) On Convergence of Sum of White Noises to Periodic Stochastic Process 10:00-11:00 *石山 健一(東京大学), 斉木 吉隆(東京大学) Unstable Periodic Orbits and Chaotic Transitions in a Macro-economic Model 11:10-12:10 細矢 祐誉(慶應義塾大学) 著作権法の有用性についての一つの疑問点 午後の部 座長 : 武隈 愼一(一橋大学) 13:30-14:30 増澤 拓也(慶應義塾大学) Computation of Cores of Strategic Games with Punishment-dominance Relations 14:30-15:30 *花薗 誠(京都大学), Huanxing Yang (Ohio State University) Collusion, Fluctuating Demand, and Price Rigidity 座長 : 浦井 憲(大阪大学) 16:00-17:00 *吉町 昭彦(大阪大学), 浦井 憲(大阪大学) 非対称情報経済のための一つの一般均衡モデル 17:00-18:00 菊田 健作(兵庫県立大学) 協力ゲームとその双対について

経済の数理解析 - Osaka Universitymath.econ.osaka-u.ac.jp › member › center › rcme › prog+abst04.pdf · 2005-02-09 · The optimal SPPE consists of a profile of price-rigidity

  • Upload
    others

  • View
    0

  • Download
    0

Embed Size (px)

Citation preview

Page 1: 経済の数理解析 - Osaka Universitymath.econ.osaka-u.ac.jp › member › center › rcme › prog+abst04.pdf · 2005-02-09 · The optimal SPPE consists of a profile of price-rigidity

経済の数理解析

京都大学数理解析研究所 研究集会

京都大学数理解析研究所の共同研究事業の一つとして、下記のように研究集会を催しますので、

ご案内申し上げます。

研究代表者 丸山 徹(慶應義塾大学)

□ 日 時 平成17年2月11日(金)-2月13日(日)

□ 場 所 京大会館 210号室

〒606-8305 京都市左京区吉田河原町15-9 ℡075-751-8311

市バス 京大正門前 下車/京都バス 荒神橋 下車

京阪電車鴨東線丸太町駅下車徒歩約10分

□ 講演時間 60分(うち質疑応答10分)

プログラム

2月11日(金) * 印:講演者

午前の部 座長 : 丸山 徹(慶應義塾大学)

9:00-10:00 加藤 寛之(慶應義塾大学)

On Convergence of Sum of White Noises to Periodic Stochastic Process

10:00-11:00 *石山 健一(東京大学), 斉木 吉隆(東京大学)

Unstable Periodic Orbits and Chaotic Transitions in a Macro-economic

Model

11:10-12:10 細矢 祐誉(慶應義塾大学)

著作権法の有用性についての一つの疑問点

午後の部 座長 : 武隈 愼一(一橋大学)

13:30-14:30 増澤 拓也(慶應義塾大学)

Computation of Cores of Strategic Games with Punishment-dominance

Relations

14:30-15:30 *花薗 誠(京都大学), Huanxing Yang (Ohio State University)

Collusion, Fluctuating Demand, and Price Rigidity

座長 : 浦井 憲(大阪大学)

16:00-17:00 *吉町 昭彦(大阪大学), 浦井 憲(大阪大学)

非対称情報経済のための一つの一般均衡モデル

17:00-18:00 菊田 健作(兵庫県立大学)

協力ゲームとその双対について

Page 2: 経済の数理解析 - Osaka Universitymath.econ.osaka-u.ac.jp › member › center › rcme › prog+abst04.pdf · 2005-02-09 · The optimal SPPE consists of a profile of price-rigidity

2月12日(土)

午前の部 座長 : 前田 隆(金沢大学)

9:00-10:00 *横田 耕祐(大阪大学), 浦井 憲(大阪大学)

抽象凸構造と一般双対構造ならびに不動点定理

10:00-11:00 高橋 渉(東京工業大学)

Fixed Point Theorems for Nonlinear Operators and Their Applications

11:10-12:10 岩本 誠一(九州大学)

黄金最適解の交差双対

午後の部 座長 : 楠岡 成雄(東京大学)

13:30-14:30 中山 季之(三菱証券)

Support Theorem と金利モデルの整合性

14:30-15:30 室井 芳史(東京大学)

ノックアウト条項の付いたルックバックオプションの評価について

座長 : 国友 直人(東京大学)

16:00-17:00 新井 拓児(東京理科大学)

不連続株価過程に対するmean-variance hedging

17:00-18:00 *上村 昌司(一橋大学), 本多 俊毅(一橋大学)

On the Verification Theorem of Continuous-time Optimal Portfolio

Problems with Stochastic Market Price of Risk

18:10- 懇親会

場所: 京大会館 211号室

2月13日(日)

午前の部 座長 : 足立 英之(流通科学大学)

9:00-10:00 春山 鉄源(神戸大学)

R&D Policy in a Volatile Economy

10:00-11:00 中村 保(神戸大学)

A Short-Run and Long-run Analysis of a Quasi-stochastic Monetary

Economy(Tentative)

11:10-12:10 武隈 愼一(一橋大学)

Rational Expectations and the Modigliani-Miller Theorem

午後の部 座長 : 丸山 徹(慶應義塾大学)

13:30-14:30 *尾崎 裕之(東北大学), 西村 清彦(東京大学)

Multiple-Belief Rational-expectation Equilibria in OLG Models with

Ambiguity

座長 : 時政 勗(広島修道大学)

14:30-15:30 藤田 敏之(九州大学)

不確実性下での国際環境協定の設計

15:40-16:40 横尾 昌紀(岡山大学)

Intersectoral knowledge spillovers and the occurrence of chaos

Page 3: 経済の数理解析 - Osaka Universitymath.econ.osaka-u.ac.jp › member › center › rcme › prog+abst04.pdf · 2005-02-09 · The optimal SPPE consists of a profile of price-rigidity

On convergence of sum of white noises toperiodic stochastic process

Hiroyuki KATOGraduate School of Economics,

Keio University

Abstract

This paper gives a proof of the convergence of a weighted linearsum of white noises to a periodic stochastic process. This problem isrooted in the classical problem on macroeconomic dynamics, especiallybusiness cycle theory that the source of energy which maintains theeconomic cycles are erratic shocks.

In order to solve this problem, we represent the covariance of thesum of white noises as a characteristic function by making use of aspectral measure and show that the spectral measure weakly convergesto a measure by which the periodic stochastic process can be repre-sented.

Page 4: 経済の数理解析 - Osaka Universitymath.econ.osaka-u.ac.jp › member › center › rcme › prog+abst04.pdf · 2005-02-09 · The optimal SPPE consists of a profile of price-rigidity

������������ ��� ����������������������������������� �����"!#��� ���������������$�%�'&(������*),+-������./����&(���� 0�

132*46587:9<;>=@?A=�BC:D 2FEG=@9IH�9KJL@LNMPORQ:SUT = S ;WVYX@X@Z

[ \@]6^`_�aRbc^d OfegT ? ORS 9ihj=@kik<;ml e:n ?A= e ; T:e 5poq= Q k Osr0ORS 9it n 9Ihs5Ut@k T oq9it e 5

O ? QPOjn:nGOjn 9 e =uhv7:=@tRow9Ihx=yoUo S =@h$owt S 9 e =z?A=@h S tR{ O hRt e t@?A9ihx| S tc}�oq7h�;0hjk O ?At n:O k~tY�0o6}Nt�hjt T:e o S 9 O 5�}�9<oq7 n 9i� OjSUORe oAl:5UhR=@k r t@ki9ihj9 O 5U2M 9 S 5po�} O ��tgh T 5�t e =�5 rPO hj9i=@k*o6; r0O t@�Foq7 OAT:e 5poq= Q k OfrPOjS 9it n 9Ih58t@k T oq9it e 582�4Ko�9I5uhRt e l S ? ORn oq7G=Ro�5U9i? r k O 5po SUT h�o T:S8O t@�~hv7:=@tRow9IhQPO 7G=v�09it S�S8Ojr:S8O 5 Oje o ORn�Q ;�oq7 O ?At n:O k�9i5�� T =@ki9<oq=Row9�� O k<;�= e:n� T = e ow9�ow=Roq9<� O k�; SUO kI=Ro Ojn owtsoq7 O�T:e 5�oq= Q k O�rPOjS 9It n 9ih�5UtYk T ow9It e 5U2d O#r t@9 e o�t T o�oq7:=Ro�ow7 O 5po SUT h�o T:S8O tY��=Whv7:=@tyoq9ih�58t@k T oq9it e 9I5n 9I585Ut@k<� ORn 9 e oqt#=#hjki=@585�t@�l e 9<o O�T:e 5�oq= Q k O�rPOjS 9It n 9ih�5Ut@k T oq9it e 5r 9IhvH Ojn t T o�=@?At e |�=�kI= S | O�e�T ? Q0ORS t@� rPOjS 9It n 9ih�5UtYk T ow9It e 5U2E O hjt e:n } O 5U7Gtv},oq7 O hRt SUS8O 5 r t eGn:Oje h O 5 Q0O o6} OROje hv7G=@tRoq9ih�o S = e {589�ow9It e 53= e:n�T:e 5pow= Q k OAr0ORS 9it n 9ihAt SUQ 9�ow5U2#4 e oq7:9i53hjt e o O�� o�9<of9I5O 5U5 Oje oq9i=@k���t S ow7 O�T:e 5pow= Q k O�rPOjS 9It n 9ih�58t@k T oq9it e 5 owt QPO�O ? QPOjn {nGOjn 9 e oq7 O hv7:=YtRoq9ih3=RoUo S =@h�oqt S 2

�`�p���*�6�<�i�j�< K¡c�<¢6�6���y£Y�* 6¤ ¥c��¦I§6¨$¢6§y¤ �6©8¤ ª¬«­ �p���*�6�<�i�@ K�6�<¨$�i£Y�* 6¤ ¥c��¦I§6¨$¢6§y¤ �6©8¤ ª¬«¯®c°:��«c�6±I¦I���p²$¦>§6³Y´x�6¦i¡c���*�6¦i�<©��6�:µ¶©��<��²c©��� 6®¸·Y±K�6¹c¥c�6¦i�µ¶©K¡c§8§6�§6³g´x�6¦I¡c���*�6¦i�<©��6�*µp©p�<��²c©��� 6®»ºg¡U�~¼�²c�<½6��±K K�<¦¾¢ §6³Gºq§6¨$¢6§y®»¿6��À6��Á�Â>§6�*�6Ãc�y®¯´x�pÄ6¥U±K§y®'ºq§6¨8¢6§�ÁKÅ6¿6�À6ÆyÁKÇy®8È��6«c�6²

É

Page 5: 経済の数理解析 - Osaka Universitymath.econ.osaka-u.ac.jp › member › center › rcme › prog+abst04.pdf · 2005-02-09 · The optimal SPPE consists of a profile of price-rigidity

A Question about Usefulness ofCopyright Protection

Yuhki HosoyaGraduate School of Economics, Keio University.

Febrruary, 2005

Abstract

This paper treats the problem of copyright protection. Many worksconsider that strong copyright protection increases producer surplusand decreases consumer surplus, but we argue that it is doubtful. Weuse a repeated game to analyze this market, and show that to weakencopyright protection may improve economic welfare in the sense ofPareto efficiency.

1

Page 6: 経済の数理解析 - Osaka Universitymath.econ.osaka-u.ac.jp › member › center › rcme › prog+abst04.pdf · 2005-02-09 · The optimal SPPE consists of a profile of price-rigidity

Computation of Cores of Strategic Games with

Punishment-Dominance Relations (要約)

増澤 拓也

慶応義塾 大学大学院経済学研究科 博士課程

この論文では、Masuzawa (2003) で定義された、公共財の供給ゲーム、クールノー寡占市

場ゲームなどを含むような、ゲームのクラスに対して、コア(α-core)に含まれるような戦略

の組を計算する O(n3 maxmi) の アルゴリズムを与える。ここで、n はプレイヤーの数、mi

は プレイヤー i が選ぶ戦略の候補の個数である。n人協力ゲームの解概念は、(1) 考慮される

提携の数が多い (2)提携が選択するとされる戦略の組が多い等の理由から、一般には戦略の

調整が困難になることが予想されるが、この特殊なクラスではこの問題がある程度解決でき

ることを示す。ここで示される方法は、Greenberg (1985), Peleg (1986) らによって論じられ

た縮小ゲームの性質を利用している。

このアルゴリズムは、初期条件を変化させることによって、さまざまな性質をもつコアに含

まれる戦略を探すことができる。第一に、対称性を持つような戦略の組を探すことができる。

ここで、対称な戦略の組とは、同等なプレイヤーは、同等な行動をとるような戦略の組であ

る。このような性質は、規範的な観点から望ましいと考えられる。ここで与えるアルゴリズム

は、どのプレイヤーの組が同等であるのかを、先験的な情報として必要としないことに注意し

なければならない。

第二に、任意に与えられた達成可能なコアに含まれない利得ベクトルに対して、それをコア

に含まれる利得ベクトルを通じて支配するような提携を探すことができる。このような提携と

利得ベクトルの存在自体は、より大きなゲームのクラスにおいて Peleg (1986) によって示さ

れているが、コアが唯一の vNM安定集合であることを意味している。我々のクラスのゲーム

においては、vNM安定集合で想定されている行動の効率的な計算も可能であることがわかる。

Reference

Greenberg, J., 1985. Cores of Convex Games without Side Paymants. Math.Oper.Res. 10,

523-525.

Masuzawa, T., 2003. Punishment Strategies Make the α-Coalitional Game Ordinally Con-

vex and Balanced. Int.J.Game Theory 32, 479-483.

Peleg, B., 1986. A Proof that the Core of an Ordinal Convex Game is a von Neumann-

Morgenstern Solution. Math.Soc.Sci. 11, 83-87.

1

Page 7: 経済の数理解析 - Osaka Universitymath.econ.osaka-u.ac.jp › member › center › rcme › prog+abst04.pdf · 2005-02-09 · The optimal SPPE consists of a profile of price-rigidity

Title: Collusion, Fluctuating Demand, and Price Rigidity Authors: Makoto Hanazono (Kyoto U) and Huanxing Yang (Ohio State U)

Abstract We study an infinitely repeated Bertrand game in which an i.i.d. demand shock occurs in each period. Each firm receives a private signal about the demand shock at the beginning of each period. At the end of each period, information about both the underlying demand shock and the rivals' prices becomes public. A firm's pricing schedule can be either a sorting scheme, in which its price depends on its private signal, or a price-rigidity scheme, in which the firm charges the same price regardless of its private signal. We consider the optimal symmetric perfect public equilibrium (SPPE). The optimal SPPE consists of a profile of price-rigidity schemes if the accuracy of the private signals is low. Moreover, the lower the variance of the demand shock, the more likely that a price-rigidity scheme is optimal. These results contribute to our understanding of which industries, and under what conditions, should exhibit rigid prices.

Page 8: 経済の数理解析 - Osaka Universitymath.econ.osaka-u.ac.jp › member › center › rcme › prog+abst04.pdf · 2005-02-09 · The optimal SPPE consists of a profile of price-rigidity

非対称情報経済のための一つの一般均衡モデル

Ken URAI∗

and

Akihiko YOSHIMACHI†

February 20, 2003

(Revised June 4, 2004 )

Abstract

本論文では, 非対称情報下での一般均衡モデルを構築する. ここで非対称情報は,標準的な一般均衡

モデルの自然な拡張としての “pseudo” テクノロジーおよびその下での取引,言うなれば架空の数

量を含む取引として描かれている.

Keywords : General Equilibrium Model, Asymmetric Information, Pseudo-technology, Ad-

verse Selection.

JEL classification: C62; D51; D82

∗Faculty of Economics, Osaka University, Machikaneyama, Toyonaka, Osaka 560-0043, JAPAN. (Internet e-mailaddress: [email protected])

†Graduate School of Economics, Osaka University, Machikaneyama, Toyonaka, Osaka 560-0043, JAPAN. (Internete-mail address: [email protected])

Page 9: 経済の数理解析 - Osaka Universitymath.econ.osaka-u.ac.jp › member › center › rcme › prog+abst04.pdf · 2005-02-09 · The optimal SPPE consists of a profile of price-rigidity

協力ゲームとその双対について

兵庫県立大学  菊田 健作

Many people examined relations between a cooperative game with transferable util-

ity and its dual, or relations between solutions for a game and its dual. Shapley

(1962) outlines theoretical structures of simple games, and gives beautiful properties

of simple games and compound simple games with respect to their dual games. A

solution for games is sometimes called to satisfy the duality if the solutions for a

game and its dual coincide. The Shapley value (Shapley, 1953) and the prenucleolus

(Schmeidler, 1969) are single-valued solutions. It is well-known that the Shapley

value satisfies the duality. On the other hand, it is easy to find a game where the

prenucleolus of a game is different from the prenucleolus of the dual game of it. Is

there another duality which is satisfied by the prenucleolus ?

In this paper we propose another kind of duality of solutions which is satisfied

by the prenucleolus. This duality is, however, satisfied by other solutions as well

as the Shapley value. So it is still a problem to find a duality-like property which

distinguishes the prenucleolus from the Shapley value. This duality is at least math-

ematically an extension of the self-duality of rules in bankruptcy problems. We call

this duality a twisted duality since there may be other extensions of the self-duality.

The Shapley value and the prenucleolus have been alternatively defined axiomati-

cally by many people. One remarkable axiom is the reduced game property, or the

consistency. We propose a property of solutions by combining the twisted duality

with reduced game property, and we characterize some solutions by a set of axioms

including this property.

Page 10: 経済の数理解析 - Osaka Universitymath.econ.osaka-u.ac.jp › member › center › rcme › prog+abst04.pdf · 2005-02-09 · The optimal SPPE consists of a profile of price-rigidity

抽象凸構造と一般双対構造ならびに不動点定理

浦井 憲

横田耕祐

概 要

線形構造から定義される凸概念の拡張である抽象凸構造をできる限り一般的な形で定式化し、同じく線形空間における双対構造を拡張した一般双対構造の定義を与え、それらの下で タイプの不動点定理、角谷タイプの不動点定理の成立を示す。両タイプの不動点定理はこの一般化された構造の下で統合的に取り扱われ、その違いが不動点性を成立させる空間の位相的性質として特徴付けられる。

Page 11: 経済の数理解析 - Osaka Universitymath.econ.osaka-u.ac.jp › member › center › rcme › prog+abst04.pdf · 2005-02-09 · The optimal SPPE consists of a profile of price-rigidity

Name: Wataru Takahashi Title of Talk: Fixed Point Theorems for Nonlinear Operators and Their Applications Abstract: In this talk, we discuss fixed point theorems and convergence theorems for nonlinear operators in Hilbert spaces and Banach spaces, and then consider their applications. We first give fixed point theorems for nonexpansive mappings in Hilbert spaces and Banach spaces. Next, we obtain some convergence theorems for inverse-strongly-monotone mappings and nonexpansive mappings in Hilbert spaces. Using these results, we consider the problem concerning strictly pseudocontractive mappings and nonexpansive mappings, and the minimization problem of convex functions. Finally, we deal with relatively nonexpansive mappings in Banach spaces which generalize simultaneously nonexpansive mappings and resolvents of maxmal monotone mappings in Banach spaces. Then, we deal with some convergence theorems for these nonlinear operators and their applications.

Address: Department of Mathematical and Computing Sciences, Tokyo Institute of Technology, Ohokayama, Meguro-ku, Tokyo 152-8552, Japan Telephone: 03-5734-3208 Fax: 03-5734-3208 E-mail: [email protected]

Page 12: 経済の数理解析 - Osaka Universitymath.econ.osaka-u.ac.jp › member › center › rcme › prog+abst04.pdf · 2005-02-09 · The optimal SPPE consists of a profile of price-rigidity

Cross-Dual on The Golden Optimum Solution

Seiichi IWAMOTO

Department of Economic EngineeringGraduate School of Economics, Kyushu University 27

Fukuoka 812-8581, Japantel&fax. +81(92)642-2488, email: [email protected]

Abstract

This paper begins with two simple minimization problems. Itends with two interesting features.

One problem is to find the shortest distance from the originto an ellipse on the plane R2. The other problem is a by-productof solving an optimality equation (Bellman equation) in non-deterministic dynamic programming, which is recently proposedby the author. This minimization has an objective — the ratioof a positive quadratic function to a non-positive one — over theunit interval.

We associates each of two minimization problems with a re-lated maximization problem. Thus we consider two couples of(minimization and maximization) problems. Further we asso-ciate each of two couples with a cross-dual couple. Finally wehave two pairs of couples. As a total we have eight optimizationproblems, each of which is to optimize one quadratic two-variableobjective function under another quadratic constraint.

An optimum solution — optimum point and optimum value— is called Golden if both the slope and the optimum value con-stitute the Golden ratio. We show two explicit relations. Oneis the Golden optimum solution. All eight problems have theGolden optimum slutions. The other is the cross-duality. Thefirst pair has a cross 2-sum property. The second has a crossinverse property.

1

Page 13: 経済の数理解析 - Osaka Universitymath.econ.osaka-u.ac.jp › member › center › rcme › prog+abst04.pdf · 2005-02-09 · The optimal SPPE consists of a profile of price-rigidity

中山季之 (Toshiyuki NAKAYAMA)

A. 講演概要

本講演では金利モデルの一種である HJM(Heath-Jarrow-Morton)モデルの consistency problemを

解くために、support theorem(確率微分方程式の解のパスの分布のサポートを記述)と viability the-

orem(確率微分方程式の解がある部分集合に留まるための条件を与える)を拡張した結果について述

べる.

金利モデルの利用に際し,モデルがもつ初期時点のフォワード・カーブをマーケットから推定されるフォ

ワード・カーブに一致させることが多い.フォワード・カーブは一般に,有限個のパラメーターを含んだ

関数の族の中から推定される.例えば xについての関数 f(x; z1, z2, z3, z4) = z1 + z2e−z4x+z3xe

−z4x

の族からなる Nelson-Siegel familyと呼ばれるものがあり,マーケットで観測される国債等の価格に

合うように 4つのパラメーター z1, z2, z3, z4が決められる.

Musielaは HJMモデルにおいて,時点 tに観測される x期間後のフォワード・レートに関して次の

ような SPDE(確率偏微分方程式)を導出した.⎧⎨⎩dr(t, x) = ∂r∂x (t, x) dt+

³Pj σj(t, x)

R x0σj(t, u)du

´dt+

Pj σj(t, x)dBj(t)

r(0, x) = f(x)(1)

この式における σj は一般には stochasticである.初期条件は,マーケットから推定されるフォワー

ド・カーブ x 7→ f(x)であり,例えば Nelson-Siegel familyのような関数の族の中から選ばれる.

SPDE (1)のある意味での解は,無限次元ヒルベルト空間 H に値を取る確率過程 (r(t, · ))t≥0 とし

て数学的定式化がなされる.一方でマーケットから推定されるフォワード・カーブは,より狭い集合

M ⊂ Hに属することになる.例えばNelson-Siegel familyのように,マーケットから推定されるフォ

ワード・カーブの族は有限個のパラメーターで表されることを考慮し,M がH の ‘有限次元部分多様

体’として定式化できる場合がよく研究されている.マーケットから日々推定されるフォワード・カー

ブはM に属するので,モデルから生成される将来時点のフォワード・カーブもM に属するのかど

うかという問題が発生し consistency problemと呼ばれる.初期条件 r(0, x) = f(x)をみたす SPDE

(1)の解を rf (t, x)と書くことにすると (解の定式化や存在と一意性が示せたとして),consistencyの

成立は f ∈ M ならば常に P (rf (t, · ) ∈ M, ∀t ≥ 0) = 1 が成立つこととして定義される.このよ

うな問題を考える動機は,モデルの日々のパラメーター推定を安定化させることにある.

そこでM ⊂ H と SPDE(1)が与えられたとき, consistentであるかどうかを判断するための条件を

考えることが重要となる. この条件をラフに述べると, SPDE(1)を H に値をとる確率 (常)微分方程

式とみなしたとき, そのドリフトを修正したものと拡散係数が M の各点でその ‘接空間’に属するこ

とである. ここでM はある意味で ‘有限次元部分多様体’とみなせることを仮定する. このことの数

学的な定式化と証明には, Bjorkや Filpovicによる微分幾何学的アプローチがある. 本講演では H 上

の確率微分方程式に対する support theorem を SPDE(1)にも適用できるように拡張し (Nakayama

[1]), その応用として導いた Nakayama [2]について述べる. 具体的には次のような形の可分ヒルベル

ト空間上の確率微分方程式の mild solution に関する support theoremと viability theoremを構築

した.

dX(t) = AX(t)dt+ b(X(t))dt+ σ(X(t))dB(t)

ここで,ウィナー過程 B は可分ヒルベルト空間に値をとり, A は 任意の (C0)-半群の生成作用素で

ある.

B. 元となる論文

[1] Nakayama, T., Support Theorem for mild solutions of SDE’s in Hilbert spaces, Journal of

mathematical sciences, the University of Tokyo, 2004.

[2] Nakayama, T., Viability Theorem for SPDE’s including HJM framework, Journal of mathemat-

ical sciences, the University of Tokyo, 2004.

Page 14: 経済の数理解析 - Osaka Universitymath.econ.osaka-u.ac.jp › member › center › rcme › prog+abst04.pdf · 2005-02-09 · The optimal SPPE consists of a profile of price-rigidity

ノックアウト条項の付いたルックバックオプションの評価について

室井芳史 近年金融取引の複雑化からさまざまな経路依存型のオプションの価格付け問題

が研究されている。本研究では Kunitomo=Ikeda(1992)による両側にしきい値のあるようなノックアウトオプション価格の解析法の研究をもとにノックアウ

ト条項のついたルックバックオプションの価格の解析を行なった。ノックアウ

トオプションおよびルックバックオプションはともに古くから研究が行われて

きた商品であり、その二つの性質を持つ商品の解析は自然な流れでありとても

興味深いものと思う。ルックバックオプションはあらかじめ決められた期間の

原証券の価格の最大値または最小値を原資とするような派生商品であり、値段

が高いことが予想される。そこで、ノックアウト条項をつけることでオプショ

ンの価格を下げることで買い手の負担を下げ、売り手の支払いリスクも減らす

ことができることに目を付けた。また、実務的にも経路依存型派生商品の価格

つけとしての価値のみならず、変額年金等の価格付け等でも大きな意味合いを

持つものではないかと思う。また、このような商品の価格は閉じた形の式で書

けることが分かったため、きわめて手軽に価格計算を行うことが可能になった。

ここでは、さまざまな形のノックアウト条項付きルックバックオプション(両側しきい値を持つフロートストライク型、フイックストストライク型、最大値と

最小値の差がペイオフのもの、アップアンドアウト、ダウンアンドアウトフロ

ートストライク型ルックバックオプション)について計算を試みた。

Page 15: 経済の数理解析 - Osaka Universitymath.econ.osaka-u.ac.jp › member › center › rcme › prog+abst04.pdf · 2005-02-09 · The optimal SPPE consists of a profile of price-rigidity

不連続株価過程に対するmean-variance hedging

新井拓児 (東京理科大学理工学部情報科学科)

e-mail:[email protected]

Mean-variance hedgingとは、非完備市場モデルにおける最適ヘッジの1つで、満期におけるポートフォリオの価値と条件付請求権との差の 2乗の期待値が最小になる self-financingなポートフォリオのことをいう。ここでは、満期 T > 0の非完備市場を考え、株価過程を不連続 semimartingale X、

self-financingなポートフォリオを predictable processで与える。このとき、初期資産 cを持った投資家の条件付請求権Hに対するmean-variance hedgingは最小化問題

Minimize E[(H − c − GT (ϑ))2

]over all ϑ ∈ Θ

の解である。ただし、GT (ϑ) ≡∫ T

0

ϑsdXs、Θは self-financingなポートフォ

リオの全体を表す。

株価過程 X が連続である時、Gourieroux, Laurent and Pham (1998)と Rheinlander and Schweizer (1997)がかなり一般的な設定の下で mean-variance hedgingの表現を得ている。ここでは、variance-optimal martingalemeasure(VOM)に関する 3つの条件の下、Gourierouxらが用いた change ofnumeraireの方法を用いて、彼らの結果を不連続の場合へ拡張する。連続株価過程モデルでは、条件付請求権の VOMの下での国田-渡辺の直交分解を用いたが、不連続の場合には一般に国田-渡辺の直交分解が存在しない。そこで、連続の場合の国田-渡辺の直交分解と同じ構成法で条件付請求権の分解を作り、その分解の性質を詳細に調べる。その結果、この分解は一般的には直

交性を持たないが、分解の各項が適当な可積分条件などを満たすことを示し、

連続モデルと同じ議論に持ちこむことが可能であることを示す。

講演の後半では、VOMに関する条件の考察を行う。1つはこの条件を満たす例の提示であり、もう 1つはこの条件を満たすための十分条件について触れる。

参考文献

[1] Gourieroux, C., Laurent, J.P. and Pham, H. (1998). Mean-variance hedging and numeraire. Math. Fin. 8, 179–200.

[2] Rheinlander, T. and Schweizer, M. (1997). On L2-projections ona space of stochastic integrals. Ann. Prob. 25, 1810–1831.

Page 16: 経済の数理解析 - Osaka Universitymath.econ.osaka-u.ac.jp › member › center › rcme › prog+abst04.pdf · 2005-02-09 · The optimal SPPE consists of a profile of price-rigidity

On the Verification Theorem of Continuous-Time

Optimal Portfolio Problems with Stochastic

Market Price of Risk

Toshiki Honda ∗ Shoji Kamimura †

In this paper, we study a continuous-time portfolio optimization problem inKim and Omberg (1996). We show sufficient conditions to verify that a solutionderived from the Hamilton-Jacobi-Bellman (HJB) equation is in fact an optimalsolution to the portfolio selection problem.

Since Merton’s seminal work (Merton (1969, 1971)), many studies have beendone on continuous-time portfolio optimization problems. In particular, therehas been increasing interest in finding an optimal portfolio strategy when in-vestment opportunities are stochastic, because many empirical works concludethat investment opportunities are time-varying. In general, it is difficult toderive an explicit solution for such problems. Kim and Omberg (1996) is aninteresting exception. In their model, a market price of risk is driven by anOrnstein-Uhlenbeck process. Furthermore, their model is an incomplete marketmodel. When a market is not complete, it is more difficult to find an analyticalsolution to the optimal portfolio problem.

There are two main approaches to solving the continuous-time portfolio op-timization problem. One is the stochastic control approach and the other is themartingale approach. Following Kim and Omberg (1996), we employ the formerapproach in this paper. Since the market is incomplete in our model, the mar-tingale approach is not applied directly. In the stochastic control approach, anoptimal solution is conjectured by solving the HJB equation. It is necessary toverify that the conjectured solution is in fact a solution to the original problem.However, as Korn and Kraft (2004) pointed out, the verification is often skippedsince it is very mathematically demanding. The conjectured solution from theHJB equation could be an incorrect solution to the original problem. Kim andOmberg (1996) examined the finiteness of the conjectured value function verycarefully. However, they did not provide a verification theorem. Our aim in thispaper is to derive sufficient conditions to confirm that the conjectured solutionis in fact the solution to the original problem.

∗Graduate School of International Corporate Strategy, Hitotsubashi University, 2-1-2 Hi-totsubashi, Chiyoda-ku, Tokyo 101-8439, Japan; Email: [email protected]

†Graduate School of International Corporate Strategy, Hitotsubashi University, 2-1-2 Hi-totsubashi, Chiyoda-ku, Tokyo 101-8439, Japan; Email: [email protected]

1

Page 17: 経済の数理解析 - Osaka Universitymath.econ.osaka-u.ac.jp › member › center › rcme › prog+abst04.pdf · 2005-02-09 · The optimal SPPE consists of a profile of price-rigidity

R&D Policy In A Volatile Economy∗

Tetsugen Haruyama†

Kobe University

December 2004

Abstract

The R&D-based growth literature establishes that market equilibriumis inefficient due to several types of market failures and derives optimalR&D policy. An important aspect of such normative analysis is thatsteady state is assumed, largely motivated by analytical convenience.This paper questions this steady state approach of normative analysis,introducing the possibility of endogenous cycles as long-run equilibria.Several interesting results emerge: (i) cycles arise in a very standardR&D-based model once it is framed in discrete time, (ii) those cycles areinefficient in the sense that they prevent welfare maximization, (iii) steadystate optimal R&D policy fails to eliminate cycles, and can create inef-ficient cycles, (iv) a trade-off between growth and volatility exists whenR&D subsidy is applied, (v) optimal R&D policy in a fluctuating econ-omy is oscillatory and state-dependent, which generalizes steady stateoptimal R&D policy. These results represent an important qualificationto the standard steady state welfare analysis of R&D-based models.

∗I am grateful to Takashi Kamihigashi, Kazuo Mino and participants of AESS Interna-tional Conference on Economic Theory in June 2004, Kobe COE Advanced Economic StudySeries Workshop in October 2004 and Second International Conference on Economic Theoryin Kyoto, December 2004 for their comments. I am responsible for all remaining errors.

†Formally known as Chol-Won Li. Graduate School of Economics, Kobe Univer-sity, Rokkodai, Nada-ku, Kobe 657-8501, Japan; (Tel/Fax) ++81-78-803-6807; (E-mail)[email protected] ; (Web) http://www.econ.kobe-u.ac.jp/~haruyama.

Page 18: 経済の数理解析 - Osaka Universitymath.econ.osaka-u.ac.jp › member › center › rcme › prog+abst04.pdf · 2005-02-09 · The optimal SPPE consists of a profile of price-rigidity

A short-run and long-run analysis of a quasi-stochastic monetary economy

by

Tamotsu Nakamura

Graduate School of Economics, Kobe University

Abstract

To analyze the liquidity trap in a dynamic optimization framework, most studies assume that

money has inherent utility. Instead of assuming money-in-utility, this paper considers

uninsurable idiosyncratic risks as the source of money demand to investigate the short-run

economic fluctuations as well as the long-run economic growth. Individuals face uncertainty

over the return to capital, and hence invest physical capital and money as a risk diversification.

To distinguish the effect of intertemporal substitution from that of risk-aversion, we utilize a

non-expected utility maximization approach. In the short-run, due to uncertainty, the economy

may falls into the liquidity trap in which an increase in money supply does not push down the

interest rate because the money demand based on precautionary motives absorb all the newly

issued money. In the long-run, there exists the optimal growth rate of money supply, which

depends not only on the degree of risk-aversion but also on the elasticity of intertemporal

substitution.

JEL classification: E21, E22, E31, E51

Key words: Uncertainty, Money, Non-expected utility maximization, Liquidity trap, Growth

Page 19: 経済の数理解析 - Osaka Universitymath.econ.osaka-u.ac.jp › member › center › rcme › prog+abst04.pdf · 2005-02-09 · The optimal SPPE consists of a profile of price-rigidity

Rational Expectations and the Modigliani-Miller Theorem

By Shin-Ichi Takekuma

The Graduate School of Economics Hitotsubashi University

Abstract In the paper a general model of dynamic economy is presented and the equilibrium of

rational expectations for the economy is defined. In the model we re-examine the Modigliani-Miller theorem which asserts that the value of a firm is independent of the debt-equity ratio. We show in the context of a dynamic general equilibrium model that the M-M theorem holds in a much more general framework. The validity of the theorem depends heavily on the rationality of individuals’ expectations.

In the proof of the M-M theorem, it was usually assumed that the gross returns of a firm depend only on the state of the world, since it was based on static equilibrium rather than dynamic analysis. In the dynamic economy, the profits of firms are determined in equilibrium depending on the behaviors of all economic agents, especially their expectations.

In the paper, we show that the M-M results are still valid in a dynamic equilibrium of rational expectations. Also, in a simplified model of economy where all the consumers are identical, we prove the existence of rational expectations equilibrium.

Page 20: 経済の数理解析 - Osaka Universitymath.econ.osaka-u.ac.jp › member › center › rcme › prog+abst04.pdf · 2005-02-09 · The optimal SPPE consists of a profile of price-rigidity

MULTIPLE-BELIEF RATIONAL-EXPECTATIONS

EQUILIBRIA IN OLG MODELS WITH AMBIGUITY∗

byKiyohiko G. Nishimura

Faculty of EconomicsThe University of Tokyo

and

Hiroyuki Ozaki

Faculty of EconomicsTohoku University

July 29, 2004

Abstract

In this paper, we extend the concept of rational-expectations equilibrium, from a tra-ditional single-belief framework to a multiple-belief one. In traditional framework of singlebelief, agents are supposed to know the equilibrium price “correctly.” We relax this require-ment in the framework of multiple beliefs. While agents do not have to know the equilibriumexactly, they must be correct in that the equilibrium price must be always contained in thesupport of each probability distribution they think possible. We call this equilibrium concepta multiple-belief rational-expectations equilibrium. We then show that the indeterminacyand complexity could arise even when the degree of risk aversion is moderate.

∗The work reported here is partially supported by a grant from the Economic and Social Research Insitute,the Cabinet Office, the Government of Japan.

Page 21: 経済の数理解析 - Osaka Universitymath.econ.osaka-u.ac.jp › member › center › rcme › prog+abst04.pdf · 2005-02-09 · The optimal SPPE consists of a profile of price-rigidity

不確実性下での国際環境協定の設計

九州大学大学院 経済学研究院

藤田敏之

論文要旨

本論文では,単純な国際環境協定のゲームモデルを記述し,不確実性の解消が汚染物質削減に関す

る協定の安定性に及ぼす影響について検討した.協定が安定であるとは,それを遵守せずに抜けてしま

い,ただ乗りをしようとする国が出現しないことをいう.分析の目的は参加を強制することができない

場合であっても各国が参加するインセンティヴをもつ「自己拘束的な」環境協定の成立の可能性を探る

ことである.

モデルは,プレイヤーである各国が他の国と協力して提携を形成するかどうかを判断するゲーム (第

1段階)と,戦略として汚染物質の削減量を毎期決定する非協力ゲーム (第 2段階)に分かれる.第 2段階

においてある提携に属する国はその提携全体の利得を最大化する戦略をとることが義務付けられる.プ

レイヤーは第 2段階での利得を考慮に入れて協定に参加するかどうかを決める.ここで利得は汚染削減

による便益から費用を引いたものとして定義される.ある国の汚染物質削減による便益はすべての国の

総削減量に比例し,削減費用は当該国の削減量の 2乗に比例することを仮定する.

汚染削減の便益を決定するパラメーターは不確実性をもつが,ある定められた時にその値が判明す

る (学習が起こる)ものとする.我々は先行研究 Na and Shin (1998)にもとづき,不確実性の解消 (学

習)の前に交渉を行う事前的交渉ケースと,学習の後に交渉を行う事後的交渉ケースでの協定の安定性

を分析する.Na and Shinの研究では 3国モデルが記述され,事前的交渉ケースのほうが望ましいこと

が指摘されている.本論文は Na and Shinのモデルを一般的な n国モデルに拡張しようという試みで

あり,すべての国からなる提携 (全体提携)による協定の安定性に分析の焦点をあてる.さらに Na and

Shin (1998)をはじめとする多くの先行研究では協定からある国が抜けてもその国以外の残りが引き続

き協定を保持して協力的な削減をするという仮定のもとでの安定性 (本論文では CS–安定性と呼ぶ)が論

じられているが,本論文では Chander and Tulkens (1997)にもとづく,そのような逸脱が生じた場合に

は協定自体が無効になり提携は解散しなければならないというルールのもとでの安定性 (γ–安定性)につ

いても検証した.

分析の結果,事前的交渉ケースにおいて,国の数によらずつねに γ–安定性が成り立つことと,CS–

安定性は国の数が 4以上になると必ず成り立たないことが示された.事後的交渉ケースでは,まず γ–安

定性は事前的交渉ケースと同じく国の数によらずつねに成り立つことが示され,次に CS–安定性は国の

数が 3以下のときにはつねに成り立ち,便益の大きい国から小さい国への別払いによって安定な全体提

携が実現されることが明らかになった.国の数が 4以上のときには,CS–安定性は便益パラメーターの

分散に依存し,ばらつきの程度が大きいほど安定になる可能性が高いことが例によって示された.

以上の結果により,交渉が行われるのが不確実性の解消より前か後かという要素は協定の安定性に

さほど大きな影響を及ぼさず,Na and Shinの主張は国の数が 4以上の場合には成り立たないというこ

とと,協定を抜ける国に対して他の国々が削減量を減らすという「懲罰」を与えることがルールの中に

盛り込まれることの必要性が示された.また安定な協定の実現には別払いの譲渡が不可欠であり,汚染

物質削減に関する技術援助や資金援助というルールづくりが重要であることもわかった.

Page 22: 経済の数理解析 - Osaka Universitymath.econ.osaka-u.ac.jp › member › center › rcme › prog+abst04.pdf · 2005-02-09 · The optimal SPPE consists of a profile of price-rigidity

Intersectoral Knowledge Spillovers and

the Occurrence of Chaos

Masanori YokooDepartment of Economics, Okayama University

email: [email protected]

abstract

As an extension of the previous works by Ishida and Yokoo (2004,2005), the present paper proposes a dynamic multi-sector economicmodel in which the payoff of each firm in a sector depends not onlyon the aggregate state of knowledge of its own sector but also onthose of the other sectors. Upon entering the market, each new firmfaces a binary choice problem: to innovate the project or not. Inno-vation is time-consuming but reduces the future cost of production.Under those and some additional assumptions, we first re-examinethe case in which each sector is isolated, which is a special casepreviously studied by Ishida and Yokoo (2004, 2005). We then cou-ple these sectors with each other by taking account of knowledgespillovers across sectors. The resulting economic model is describedby a higher-dimensional discrete-time dynamical system. We showthat the model can exhibit chaotic behavior, by finding a homoclinicpoint associated with a repeller-type steady state.