A Consistent Nonparametric Test for Causality in Quantile

Size: px
Start display at page:

Download "A Consistent Nonparametric Test for Causality in Quantile"

Transcription

1 SFB 649 Discussion Paper 8-7 A Consisen Nonparaeric es for Causaliy in Quanile Kiho Jeong* Wolfgang Härdle** * Kyungpook Naional Universiy Daegu, Korea ** Hubold-Universiä zu Berlin, Gerany SFB E C O N O M I C R I S K B E R L I N his research was suppored by he Deusche Forschungsgeeinschaf hrough he SFB 649 "Econoic Risk". hp://sfb649.wiwi.hu-berlin.de ISSN SFB 649, Hubold-Universiä zu Berlin Spandauer Sraße, D-78 Berlin

2 A Consisen Nonparaeric es for Causaliy in Quanile Kiho Jeong School of Econoics and rade Kyungpook Naional Universiy Daegu 7-7, Korea Eail: Wolfgang Karl Härdle CASE - Cener for Applied Saisics and Econoics Hubold-Universiä zu Berlin Wirschafswissenschafliche Fakulä Spandauer Srasse, 78 Berlin, Gerany Eail: haerdle@wiwi.hu-berlin.de Absac his paper proposes a nonparaeric es of causaliy in quanile. Zheng (998) has proposed an idea o reduce he proble of esing a quanile resricion o a proble of esing a paricular ype of ean resricion in independen daa. We exend Zheng s approach o he case of dependen daa, paricularly o he es of Granger causaliy in quanile. he proposed es saisic is shown o have a second-order degenerae U-saisic as a leading er under he null hypohesis. Using he resul on he asypoic noral disribuion for a general second order degenerae U-saisics wih weakly dependen daa of Fan and Li (996), we esablish he asypoic disribuion of he es saisic for causaliy in quanile under β-ixing (absoluely regular) process. Key Words: Granger Causaliy, Quanile, Nonparaeric es JEL classificaion: C4, C5 We hank Jürgen Franke for his Malab code o copue a nonparaeric kernel esiaor of condiional quanile. he research was conduced while Jeong was visiing CASE-Cener for Applied Saisics and Econoics, Hubold-Universiä zu Berlin in suers of 5 and 7. Jeong is graeful for heir hospialiy during he visi. Jeong s work was suppored by he Korean Research Foundaion Gran funded by he Korean Governen (MOEHRD) (KRF-6-B) and Härdle s work was suppored by he Deusche Forschungsgeeinschaf hrough he SFB 649 "Econoic Risk".

3 . Inroducion Wheher oveens in one econoic variable cause reacions in anoher variable is an iporan issue in econoic policy and also for he financial invesen decisions. A fraework for invesigaing causaliy has been developed by Granger (969). esing for Granger causaliy beween econoic ie series has been sudied inensively in epirical acroeconoics and epirical finance. he ajoriy of research resuls have been obained in he conex of Granger causaliy in he condiional ean. he condiional ean, hough, is a quesionable eleen of analysis if he disribuions of he variables involved are non-ellipic or fa ailed as o be expeced wih financial reurns. he fixaion of causaliy analysis on he ean igh resul in any unclear resuls on Granger causaliy. Also, he condiional ean arges on an overall suary for he condiional disribuion. A ail area causal relaion ay be quie differen o ha of he cener of he disribuion. Lee and Yang (7) explore oney-incoe Granger causaliy in he condiional quanile by using paraeric quanile regression and find ha Granger causaliy is significan in ail quaniles, while i is no significan in he cener of he disribuion. his paper invesigaes Granger causaliy in he condiional quanile. I is well known ha he condiional quanile is insensiive o oulying observaions and a collecion of condiional quaniles can characerize he enire condiional disribuion. Based on he kernel ehod, we propose a nonparaeric es for Granger causaliy in quanile. esing condiional quanile resricions by nonparaeric esiaion echniques in dependen daa siuaions has no been considered in he lieraure before. his paper herefore inends o fill his lieraure gap. Recenly, he proble of esing he condiional ean resricions using nonparaeric esiaion echniques has been acively exended fro independen daa o dependen daa. Aong he relaed work, only he esing procedures of Fan and Li (999) and Li (999) are consisen and have he sandard asypoic disribuions of he es saisics. For he general hypohesis esing proble of he for E( ε z) = a.e., where ε and z are he regression error er and he vecor of regressors respecively, Fan and Li (999) and Li (999) all consider he disance easure of J = E[ ε E( ε z) f( z)] o consruc kernel-based consisen es procedures. For he advanages of using disance easure J in kernel-based

4 esing procedures, see Li and Wang (998) and Hsiao and Li (). A feasible es saisic based on he easure J has a second order degenerae U-saisics as he leading er under he null hypohesis. Generalizing Hall s (984) resul for independen daa, Fan and Li (999) esablish he asypoic noral disribuion for a general second order degenerae U-saisics wih dependen daa. All he resuls saed above on esing ean resricions are however irrelevan when esing quanile resricions. Zheng (998) proposed an idea o ransfor quanile resricions o ean resricions in independen daa. Following his idea, one can use he exising echnical resuls on esing ean resricions in esing quanile resricions. In his paper, by cobining he Zheng s idea and he resuls of Fan and Li (999) and Li (999), we derive a es saisic for Granger causaliy in quanile and esablish he asypoic noral disribuion of he proposed es saisic under he bea-ixing process. Our esing procedure can be exended o several hypoheses esing probles wih condiional quanile in dependen daa; for exaple, esing a paraeric regression funcional for, esing he insignificance of a subse of regressors, and esing seiparaeric versus nonparaeric regression odels. he paper is organized as follows. Secion presens he es saisic. Secion 3 esablishes he asypoic noral disribuion under he null hypohesis of no causaly in quanile. echnical proofs are given in Appendix.. Nonparaeric es for Granger-Causaliy in Quanile o siplify he exposiion, we assue a bivariae case, or only{ y, w } are observable. Denoe U = { y, L, yp, w, L, wq} and W = { w, L, wq}. Granger causaliy in ean (Granger, 988) is defined as (i) w does no cause y in ean wih respec o U if E( y U ) = E( y U W ) and (ii) w is a pria facie cause in ean of y wih respec o U if E( y U ) E( y U W ), Moivaed by he definiion of Granger-causaliy in ean, we define Granger causaliy in

5 quanile as () w does no cause y in quanile wih respec o U if Q ( y U ) = Q ( y U W ) and () () w is a pria facie cause in quanile of y wih respec o U if Q ( y U ) Q ( y U W ), () where Q ( y ) inf { y F( y ) } is he h( < < ) condiional quanile of y. Denoe x ( y, L, yp), z ( y, L, yp, w, L, wq), and he condiional disribuion funcion y given v by ( ) yv F y v, v ( x, z) =. Denoe Q ( v ) Q ( y v ). In his paper, F ( y v ) is assued o be absoluely coninuous in y for alos all ( x, z) yv v=. hen we have { } =, v= ( x, z) F Q ( v ) v yv and fro he definiions () and (), he hypoheses o be esed are H : { Fyz Q x z } H: { Fyz Q x z } Pr ( ( ) ) = = (3) Pr ( ( ) ) = <. (4) Zheng (998) proposed an idea o reduce he proble of esing a quanile resricion o a proble of esing a paricular ype of ean resricion. he null hypohesis (3) is rue if and only if E I{ y Q ( x) z} = or I{ y Q ( x) } = + ε where E( z) ε = and I() is he indicaor funcion. here is a rich lieraure on consrucing nonparaeric ess for condiional ean resricions. Refer o Li (998) and Zheng (998) for he lis of relaed works. While various disance easures can be used o consisenly es he hypohesis (3), we consider he following disance easure, { } J E Fyz( Q ( x) z) fz( z), (5) where fz ( z ) be he arginal densiy funcion of z. Noe ha J and he equaliy holds if and only if H is rue, wih sric inequaliy holding under H. hus J can be 3

6 used as a proper candidae for consisen esing H (Li, 999, p. 4). Since { } E( ε z ) = F Q ( x ) z, we have y z { ε ε } J = E E( z ) f ( z ). (6) z he es is based on a saple analog of E{ ε E( ε z) f ( z)}. Including he densiy funcion f ( z ) is o avoid he proble of riing on he boundary of he densiy funcion, z see Powell, Sock, and Soker (989) for an analogue approach. he densiy weighed condiional expecaion E( ε z) f ( z) can be esiaed by kernel ehods z ˆ Eˆ( ε z ) fz ( z ) = Ksε s, (7) ( ) h where p q s = + is he diension of z, K K{ ( z z )/ h} s s h is a bandwidh. hen we have a saple analog of J as J s s ( ) h = s = s K ε ε z = is he kernel funcion and = Ks I{ y Q ( x )} I{ ys Q ( xs ) } ( ) h (8) he -h condiional quanile of y given x, Q ( x ), can also be esiaed by he nonparaeric kernel ehod Qˆ x = Fˆ x, (9) ( ) y x ( ) where Fˆ ( y x ) = yx s LI( y y) s s s L s () is he Nadaraya-Wason kernel esiaor of Fyx ( y x ) wih he kernel funcion of L x L x a s s = esiaed as: and he bandwidh paraeer of a. he unknown error ε can be 4

7 { ˆ } ˆ ε I y Q ( x ). () Replacing ε by ˆε, we have a kernel-based feasible es saisic of J, Jˆ K ˆˆ ε ε s s ( ) h = s = s { ( )} { ( )} = K ˆ ˆ s I y Q x I ys Q xs ( ) h () 3. he Liiing Disribuions of he es Saisic wo exising works are useful in deriving he liiing disribuion of he es saisic; one is heore.3 of Franke and Mwia (3) on he unifor convergence rae of he nonparaeric kernel esiaor of condiional quanile; anoher is Lea. of Li (999) on he asypoic disribuion of a second-order degenerae U-saisic, which is derived fro heore. of Fan and Li (999). We resae hese resuls in leas below for ease of reference. Lea (Franke and Mwia) Suppose Condiions (A)(v)-(vii) and (A)(iii) of Appendix hold. he bandwidh sequence is such ha a= o() and S% a s for p = ( log ) soe s. Le S = a + S %. hen for he nonparaeric kernel esiaor of / condiional quanile of Qˆ ( x ) of equaion (9), we have ˆ sup Q( x) Q( x) = O( S ) + O p x G a a.s. (3) Lea (Li / Fan and Li) Le L ε z = (, ) be a sricly saionary process ha saisfies he condiion (A)(i)-(iv) of Appendix, ε R and z R, () K be he kernel funcion wih h being he soohing paraeer ha saisfies he condiion (A)(i)-(ii) of Appendix. Define σ ( z) = E[ ε z = z] and (4) ε 5

8 J K ε ε s s ( ) h = s (5) hen h J N(, σ ) in disribuion, (6) / 4 where σ E { σ ε ( z) fz( z) }{ K ( u) du} = and f () is he arginal densiy funcion of z z. echnical condiions required o derive he asypoic disribuion of J ˆ are given in Appendix, which are adoped fro Li (999) and Franke and Mwia (3). In he assupions we use he definiions of Robinson (988) for he class of kernel funcions ϒ v and he class of funcions A v, defined in Appendix. Condiions (A)(i)-(iv) and (A)(i)-(ii) are adoped fro condiion (A) and (A) of Li (999), which are used o derive he asypoic noral disribuion of a second-order degenerae U-saisic. Condiions (A)(v)-(vii) and (A)(iii) are condiions (A), (A), (B), (B), (C) and (C) of Franke and Mwia (3), which are required o ge he unifor convergence rae of nonparaeric kernel esiaor of condiional quanile wih ixing daa. Finally Condiions (A)(iv)-(v) are adoped fro condiions of Lea of Yoshihara (976), which are required o ge he asypoic equivalence of nondegenerae U-saisic and is projecion under he β -ixing process. We consider esing for local deparures fro he null ha converge o he null a he rae h / /4 H :. More precisely we consider he sequence of local alernaives: Fyz { Q ( x) dl( z) z} + =, (7) where d h / /4 = and he funcion () l and is firs-derivaives are bounded. heore. Assue he condiions (A) and (A). hen (i) Under he null hypohesis (3), h J ˆ N(, σ ) in disribuion, where / 6

9 { z } 4 σ = E σ ε ( z ) f ( z ) K ( u) du and σ ( ) ( ε z = E ε z) = ( ). (ii) under he null hypohesis (3), ˆ σ ( ) K s ( ) h 4 = E ( z ) f ( z ) K ( u) du. esiaor of σ { σ ε z } s is a consisen (iii) under he alernaive hypohesis (4), ˆ p {[ y z( ( ) ) ] z( )} J E F Q x z f z >. (iv) under he local alernaives (7), h J ˆ N( μ, σ ) in disribuion, where / { } μ = E fyz Q ( z) z l ( z) fz( z). heore generalizes he resuls of Zheng (998) of independen daa o he weakly dependen daa case. A deailed proof of heore is given in he Appendix. he ain difficuly in deriving he asypoic disribuion of he saisic defined in equaion () is ha a nonparaeric quanile esiaor is included in he indicaor funcion which is no differeniable wih respec o he quanile arguen and hus prevens aking a aylor expansion around he rue condiional quanile Q ( x ). o circuven he proble, Zheng (989) appealed o he work of Sheran (994) on unifor convergence of U-saisics indexed by paraeers. In his paper, we bound he es saisic by he saisics in which he nonparaeric quanile esiaor in he indicaor funcion is replaced wih sus of he rue condiional quanile and upper and lower bounds consisen wih unifor convergence rae of he nonparaeric quanile esiaor, ( y Q ( x ) C ) and ( y Q ( x ) + C ). An iporan furher sep is o show ha he differences of he ideal es saisic J given in equaion (8) and he saisics having he indicaor funcions obained fro he firs sep saed above is asypoically negligible. We ay direcly show ha he second oens of he differences are asypoically negligible by using he resul of Yoshihara (976) on he bound of oens of U-saisics for absoluely regular processes. However, i is edious o ge bounds on he second oens wih dependen daa. In he proof we insead use he fac ha differences are second-order degenerae U-saisics. hus by using he resul on he asypoic noral disribuion of he second-order degenerae U-saisic of Fan and Li 7

10 (999), we can derive he asypoic variance which is based on he i.i.d. sequence having he sae arginal disribuions as weakly dependen variables in he es saisic. Wih his lile rick we only need o show ha he asypoic variance is o () in an i.i.d. siuaion. For deails refer o he Appendix. 4. Conclusion his paper has provided a consisen es for Granger-causaliy in quanile. he es can be exended o esing condiional quanile resricions wih dependen daa; for exaple, esing isspecificaion es, esing he insignificance of a subse of regressors, esing soe seiparaeric versus nonparaeric odels, all in quanile regression odels. 8

11 Appendix Here we collec all required assupions o esablish he resuls of heore. (A) (i) { y, w } is sricly saionary and absoluely regular wih ixing coefficiens β ( τ) = O( ρ τ ) for soe < ρ <. (ii) For soe ineger v, f y, f z, and f x all are bounded and belong o A v (see D). (iii) wih probabiliy one, E z z =. [ ε μ ( ), μ ( )] E 4 ε +η < + ξ i i il E εε ε l < L for soe arbirarily sall η > and ξ >, where l 4 is and an ineger, i j 4 and l i j 8. σε ( z) = E( ε z), j= μ z = E z = z all ( ) 4 ε 4 ε saisfy soe Lipschiz condiions: au ( + v) au ( ) Du ( ) v wih for soe sall ' f τ,, τ l (iv) Le ( ) hen ( ) η >, where () σ ( ), μ ( ) a ε ε4 =. E D( z) K be he join probabiliy densiy funcion of ( zτ zτ ) f τ,, τ l,, l +η ' < K ( l 3). K is bounded and saisfies a Lipschiz condiion: ( +, +, K + ) (,, K ) ( ) f z u z u z u f z z z, K, l l l, K, l l τ τ τ τ D,,,, τ K τ z K z l l u, where K τ ( ) is inegrable and saisfies he condiion ha (,,,, l ) D τ,, l (,, ) (,, ), K, l l, K, l Dτ τ z K z fτ τ z K z dz < M < for soe ξ >. l D z z z ξ τ M K τ K < <, l (v) For any yx, saisfying < Fyx ( y x) < and fx ( x ) > ; for fixed y, he condiional disribuion funcion F yx and he condiional densiy funcion f yx belong o A 3 ; fyx ( Q ( x) x) > for all x ; f yx is uniforly bounded in x and y by c f, say. (vi) For soe copac se G, here are ε >, γ > such ha f x γ for all x in he ε -neighborhood { x x u ε, u G } < of G ; For he copac se G and soe 9

12 copac neighborhood Θ of, he se { v Q ( x) x G, } Θ = = + μ μ Θ is copac and for soe consan c >, f ( v x) c for all x G, v Θ. (vii) here is an yx increasing sequence s of posiive inegers such ha for soe finie A, s /(3 ) ( s ) A s β, s for all. (A) (i) we use produc kernels for boh L( ) and K ( ), le l and k be heir corresponding univariae kernel which is bounded and syeric, hen l() is non-negaive, l() ϒ, v k() is non-negaive and k() ϒ. (ii) h ' = O( α ) for soe α ' (7/8) < <. (iii) a = o() and % p = ( log ) for soe S a s s (iv) here exiss a posiive nuber δ such ha for r = + δ and a generic nuber M r z z K dfz( z) dfz( z) M < h h and r z z E K M h h < (v) for soe δ ' ( < δ ' < δ), β ( + δ ')/ δ ' ( ) O( ) =. he following definiions are due o Robinson (988). Definiion (D) ϒ, λ is he class of even funcions k: R R saisfying where λ i uk( u) du = δ R i ( i =,,, λ ) ( ++ ) K, ku ( ) = O( + u λ ε ), for soe ε >, δ ij is he Kronecker s dela. α Definiion (D) A μ, α >, μ > is he class of funcions g: R R saisfying ha

13 g is ( d ) -ies parially differeniable for d μ d ; for soe ρ >, μ for all z, where φzρ = { y y z < ρ} sup g( y) g( z) G ( y, z) / y z D ( z) y φzρ g g ; G g = when d = ; G g is a ( d ) h degree hoogeneous polynoial in y z wih coefficiens he parial derivaives of g a z of orders hrough d when d > ; and gz ( ), is parial derivaives of order d and less, and D ( z ), has finie α h oens. g Proof of heore (i) In he proof, we use several approxiaions o J ˆ. We define he now and recall a few already defined saisics for convenience of reference. Jˆ J J J K ˆˆ ε ε s s ( ) h = s (A.) s s ( ) h = s K ε ε (A.) U s U su ( ) h = s K ε ε (A.3) L s L sl ( ) h = s where ˆ ε { ˆ I y Q ( x) } K ε ε (A.4) =, { ( )} ε = I y Q x, ε { ( )} = I y + C Q x, U { ( )} ε = I y C Q x and L C is an upper bound consisen wih he unifor convergence rae of he nonparaeric esiaor of condiional quanile given in equaion (3). he proof of heore (i) consiss of hree seps. Sep. Asypoic noraliy: h J N(, σ ), (A.5) /

14 where σ E { ( ) f ( z )}{ K ( u) du} = under he null. Sep. Condiional asypoic equivalence: Suppose ha boh h / ( J J ) = o () and U p h / ( J J ) = o (). U p hen h ( J J ) = o (). (A.6) / ˆ p Sep 3. Asypoic equivalence: h / ( J J ) = o () and U p h / ( J J ) = o (). (A.7) L p he cobinaion of Seps -3 yields heore (i). Sep : Asypoic noraliy. Since J is a degenerae U-saisic of order, he resul follows fro Lea. Sep : Condiional asypoic equivalence. he proof of Sep is oivaed by he echnique of Härdle and Soker (989) which was used in reaing riing indicaor funcion asypoically. Suppose ha he following wo saeens hold. h / ( J J ) = o () and (A.8) U p h / ( J J ) = o () (A.9) L p Denoe C as an upper bound consisen wih he unifor convergence rae of he nonparaeric esiaor of condiional quanile given in equaion (3). Suppose ha sup Qˆ ( x) Q( x) C. (A.) If inequaliy (A.3) holds, hen he following saeens also hold: { Q ( x) > y + C } { Qˆ ( x) > y } { Q ( x) > y C }, (A.-) ( Q ( x) > y + C ) ( Qˆ ( x) > y ) ( Q ( x) > y C ), (A.-) J Jˆ J, and (A.-3) U L ˆ (A.-4) / / / h ( J J) ax { h ( J JU), h ( J JL) }

15 Using (A.) and (A.-4), we have he following inequaliy; / { h J ˆ ˆ J δ Q x Q x C} / / Pr { ax { h ( J ), ( ) } > sup ˆ JU h J JL δ Q( x) Q( x) C} Pr ( ) > sup ( ) ( ), for all δ >. (A.) Invoking Lea and condiion A(iii), we have { Q ˆ x Q x C } Pr sup ( ) ( ) as. (A.3) By (A.8) and (A.9), as, we have / / { h J JU h J JL δ } Pr ax { ( ), ( ) } >, for all δ >. herefore, as, / he L.H.S. of he inequaliy (A.) { h J ˆ J δ } he L.H.S. of he inequaliy (A.). (A.4) Pr ( ) > and In suary, we have ha if boh h / ( J J ) = o () and U p h / ( J J ) = o (), U p hen h ( J J ) = o (). / ˆ p Sep 3: Asypoic equivalence. In he reaining proof, we focus on showing ha h / ( J J ) = o (), wih he proof U p of h / ( J J ) = o () being reaed siilarly. he proof of Sep 3 is close in lines L p wih he proof in Zheng (998). Denoe H (,, s g) K {( y g( x )) }{( y g( x )) } and (A.5) s s s Jg [ ] H ( sg,, ) ( ) h =. (A.6) s hen we have J J[ Q ] and JU J[ Q C ]. We decopose H (,, s g ) ino hree pars; H ( s,, g) = K {( y g( x )) F( g( x ) z )}{( y g( x )) F( g( x ) z )} s s s s s 3

16 + K {( y g( x )) F( g( x ) z )}{ F( g( x ) z ) } s s s + K{ Fgx ( ( ) z) }{ Fgx ( ( ) z) } s s s = H (,, sg) + H (,, sg) + H (,, sg) (A.7) 3 hen le J j[ g] = H j( s,, g) ( ) h =,,, 3 s i =. We will rea J [ Q ] J [ Q C ] for j =,,3 separaely. j j h J ( Q ) J ( Q C ) = o (): / [] [ ] By siple anipulaion, we have J Q J Q C ( ) ( ) p = ( ) h = s = s [ H (,, s Q ) H (,, s Q C )] = Ks Q F Q s Q Fs Q ( ) h { [ ( ) ( ) ][ ( ) ( ) ] [ ( Q C) F( Q C) ][ s( Q C) Fs( Q C) ] } (A.8) o avoid edious works o ge bounds on he second oen of J ( Q) J ( Q C ) wih dependen daa, we noe ha he R.H.S. of (A.8) is a degenerae U-saisic of order. hus we can apply Lea and have [ ] h J ( Q ) J ( Q C ) N(, ) in disribuion, (A.9) / σ where he definiion of he asypoic variance σ is based on he i.i.d. sequence having he sae arginal disribuions as weakly dependen variables in (A.8). ha is, [ (,, ) (,, )] σ = E% H s Q H s Q C, where he noaion E % is expecaion evaluaed a an i.i.d. sequence having he sae arginal disribuion as he ixing sequences in (A.8) (Fan and Li (999), p. 48). Now, o h J ( Q ) J ( Q C ) = o (), we only need o show ha he asypoic / show ha [ ] p 4

17 variance σ z is () ( ) o wih i.i.d daa. We have [ (,, ) (,, )] E% H sq H sq C { [ Q F Q ][ s Q Fs Q ] ΛE % ( ) ( ) ( ) ( ) [ ( Q C ) F( Q C )][ ( Q C ) F ( Q C )] } s s { ( ) [ ( ) ] s( ) [ s( ) ]} ΛE% F Q F Q F Q F Q { ( ) [ ( ) ] s( ) [ s( ) ]} + E% F Q C F Q C F Q C F Q C { [ ] E F(in( Q, Q C ) F( Q ) F( Q C ) [ Fs(in( Q, Q C) Fs( Q) Fs( Q C) ] } {[ ( ) ( ) ( ) ][ s( ) s( ) s( ) ]} =ΛE% F Q F Q F Q F Q F Q F Q Λ E% { F(in( Q, Q C) F( Q) F( Q C) [ ] [ Fs(in( Q, Q C) Fs( Q) Fs( Q C) ] } +Λ E% { F( Q C) F( Q C) F( Q C) [ ] [ Fs( Q C) Fs( Q C) Fs( Q C) ] } Λ E% { F(in( Q, Q C) F( Q) F( Q C) [ ] [ Fs(in( Q, Q C) Fs( Q) Fs( Q C) ] } Λ C = o(). (A.) where he las equaliy holds by he soohness of condiional disribuion funcion and is bounded firs derivaive due o Assupion (A.8). hus we have [ ] / h J Q J Q C op ( ) ( ) = () (A.) h J ( Q ) J ( Q C ) = o (): / [] [ ] Noing ha H (,, ) s Q = because of F ( Q ( x ) z ) =, we have p yz s s J Q J Q C ( ) ( ) 5

18 =J ( ) Q C z z = ( ) h h s K = s {( y Q ( x ) C ) F ( Q ( x ) C z )}{ F ( Q ( x ) C z ) } (A.) y z y z s s Denoe Sg ( ) Fg [ ]/ g. By aking a aylor expansion of F ( Q ( x ) C z ) around Q ( x s ), we have yz s s J Q J Q C ( ) ( ) z z = K {( y Q( x) C) Fy z( Q( x) C z)} ( ) h h s = s ( C ) S( Q ( x )) s = C {( y Q ( x ) C ) F ( Q ( x ) C )} S( Q ( x )) fˆ ( z ) y z s z = C ( ( )) ˆ u S Q x f ( z ), (A.3) s z = where Q is beween Q and Q C. hus we have E J Q J Q C ( ) ( ) ΛC ˆ E u f ( z ) z = { } ΛC ˆ E u f ( z ) z = ( ( ) ) = O C h, (A.4) where he firs inequaliy holds due o Assupion ()(v) and he las equaliy is derived by using Lea C.3(iii) of Li (999) ha is proved in he proof of Lea A.4(i) of Fan and Li (996c). hus, we have 6

19 [ ( ) ( )] / h J Q J Q C = O p / ( Ch ) = o p (). (A.5) h J ( Q ) J ( Q C ) = o (): / [3] [ ] 3 3 p Noing ha H (,, ) 3 s Q = because of FQ ( ( x) z) = for j= s,, we have j j J3 Q J3 Q C ( ) ( ) z z = ( ) h h s K = s { FQ ( ( x) C z) }{ FQ ( ( x) C z) } s s z z s = K ( ( )) ( ( )) CS Q x S Q xs ( ) = s h h ˆ (A.6) = C S( Q( x)) S( Q( xs)) fz( z) = hus, we have E J3 Q J3 Q C ( ) ( ) ΛC E fˆ ( z ) z = Λ C E f ( z ) +ΛC E fˆ ( z ) f ( z ) z z z = = { } Λ C Ef ( z ) +ΛC E fˆ ( z ) f ( z ) z z z = = ( ) = O C (A.7) Finally, we have [ ( ) ( )] / h J3 Q J3 Q C p / ( ) = O h C 7

20 = o p (). (A.8) By cobining (A.), (A.5) and (A.8), we have he resul of Sep 3 Proof of heore (ii) Since { z } σ = ( ) E f ( z ) K ( u) du and ˆ σ ( ) K, s ( ) h s i is enough o show ha σ Noe ha K s ( ) h s { } = E fz ( z) K ( u) du+ op() (A.9) σ is a nondegenerae U-saisic of order wih kernel z zs H( z, zs) = K h h. (A.3) Since Assupion (A)(iv)-(v) saisfy he condiions of Lea of Yoshihara (976) on he asypoic equivalence of U-saisic and is projecion under β -ixing, we have for γ = ( δ δ ') / δ '( + δ) > σ (, ) H z zs ( ) s = H ( z, z ) df ( z ) df ( z ) z z z z z p = + H ( z, z ) df ( z ) H ( z, z ) df ( z ) df ( z ) + O ( γ ) = H ( z, z ) df ( z ) df ( z ) + o () z z p z z = K dfz( z) dfz( z) op() + h h 8

21 ( ) z p (A.3) = K u du f ( z ) dz + o () he resul of heore (ii) follows fro (A.3). Proof of heore (iii) he proof of heore (iii) consiss of he wo seps. Sep. Show ha Jˆ = J + o () under he alernaive hypohesis (4). p Sep. Show ha J = J + o () under he alernaive hypohesis (4), p where J = E F Q x z f z. he cobinaion of Seps and yields {[ yz ( ( ) ) ] z( )} heore (iii). Sep : Show ha Jˆ = J + o () under he alernaive hypohesis. p We need o show ha he resuls of Sep and Sep 3 in he proof of heore (i) hold under he alernaive hypohesis. Firs, we show ha he resul of Sep in he proof of heore (i) sill holds under he alernaive hypohesis. We can show ha J ( ) () Q C = o by he sae procedures as in (A.4). hus we focus on showing ha J ( ) () Q = o p. As in he proof of heore (i), denoe Sg ( ) Fg [ ]/ g. By aking a aylor expansion of Fyz ( Q ( xs) zs) around Q ( z s ), we have J ( Q ) z zs = K {( ( )) y z( ( ) )} y Q x F Q x z ( ) = s h h SQ ( ( x, z)) s s p = {( ( )) ˆ y Q x Fy z( Q( x))} S( Q( xs, zs)) fz( z) = ( (, )) ˆ us Q xs zs fz( z), (A.3) = where Q ( x, z ) is beween Q ( x s ) and Q ( z s ). By using he sae procedures as in s s 9

22 (A.4), we have ( ) ( ) J Q = O h. (A.33) Nex, we show ha he resul of Sep 3 in he proof of heore (i) holds under he alernaive hypohesis. Since FQ ( ( x) z) for j= s, under he alernaive hypohesis, we have J3 Q J3 Q C ( ) ( ) j j z z = { ( ( ) ) }{ ( ( s) s) } ( ) h h FQ x z FQ x z s K = s z z ( ) h h s K = s { FQ ( ( x) C z) }{ FQ ( ( x) C z) } s s = { ( ( ) ) }{ ( ( ) ) } ˆ FQ x z FQ xs zs fz( z) = { ( ( ) ) }{ ( ( ) ) } ˆ FQ x C z FQ xs C zs fz( z). (A.34) = By aking a aylor expansion of F ( Q ( x ) C z ) around Q ( z ) for j =, s, we have J3 Q J3 Q C ( ) ( ) yz j j j = { ( ( ) ) } ( ( )) ˆ FQ x z CSQ xs fz( z) = + ( ( )){ ( ( ) ) } ˆ CSQ x FQ xs zs fz( z) = CSQ ˆ x SQ xs fz z = ( ( )) ( ( )) ( ). (A.35) We furher ake aylor expansion of F ( Q ( x ) z ) around Q ( z ) for j =, s and have J3 Q J3 Q C ( ) ( ) yz j j j

23 = SQ ( (, )) ( ( )) ˆ x z CSQ xs fz( z) = + CSQ ( ( )) ( (, )) ˆ x SQ xs zs fz( z) = CSQ ˆ x SQ xs fz z = ( ( )) ( ( )) ( ), (A.36) where Q ( x, z ) is beween Q ( x s ) and Q ( z s ). hen by using he sae procedures as in (A.7), we have s s 3 3 ( ) J ( Q ) J ( Q C ) = O C. (A.37) Now we have he resul of Sep for he proof of heore (iii). Sep : Show ha J = J + o () under he alernaive hypohesis. p Using (7) and unifor convergence rae of kernel regression esiaor under β -ixing process, we have J = s s ( ) h = s K ε ε = E ˆ( ε ) ˆ ( ) z fz z ε = = = E( ε z ) f ( z ) ε z = { E ˆ( z ) f ( z ) E ( z ) f ( z ) } + ˆ ε z ε z ε = E( ε z) fz( z) ε + op() = [ ε ε ] = E E( z ) f ( z ) + o () z p = J + o p () (A.38)

24

25 References Cai, Z. (), Regression quaniles for ie series, Econoeric heory 8, Chen, X. and Y. Fan (999), Consisen hypohesis esing in seiparaeric and nonparaeric odels for econoeric ie series, Journal of Econoerics 9, Fan, Y. and Q. Li (999), Cenral lii heore for degenerae U-saisics of absoluely regular processes wih applicaions o odel specificaion ess, Journal of Nonparaeric Saisics, Franke, J. and P. Mwia (3), Nonparaeric esiaes for condiional quaniles of ie series, Repor in Wirschafsaheaik 87, Universiy of Kaiserslauern. Granger, C.W.J. (969), Invesigaing causal relaions by econoeric odels and cross-specral ehods, Econoerica 37, Granger, C.W.J. (988), Soe recen developens in a concep of causaliy, Journal of Econoerics 39, 99-. Györfi, L., W. Härdle, P. Sarda, and P. Vieu (989), Nonparaeric Curve Esiaion fro ie Series, Springer-Verlag: Berlin. Hall, P. (984), Cenral lii heore for inegraed square error of ulivariae nonparaeric densiy esiaors, Journal of Mulivariae Analysis 4, -6. Hansen, B.E. (4), Unifor convergence raes for kernel esiaion wih dependen daa, working paper, Universiy of Wisconsin-Madison, Härdle, W. and. Soker (989), Invesigaing sooh uliple regression by he ehod of average derivaives, Journal of he Aerican Saisical Associaion 84, Hsiao, C. and Q. Li (), A consisen es for condiional heeroskedasiciy in ie-series regression odels, Econoeric heory 7, 88-. Lee,. and W. Yang (7), Money-incoe Granger-causaliy in quaniles, unpublished paper, Universiy of California, Riverside Li, Q. (99), Consisen odel specificaion ess for ie series econoeric odels, Journal of Econoerics 9, -47. Li, Q. and S. Wang (998), A siple consisen boosrap es for a paraeric regression funcional for, Journal of Econoerics 87, Mwia, P. (3), Seiparaeric esiaion of condiional quaniles for ie series wih applicaions in finance, Ph D hesis, Universiy of Kaiserlauern. Powell, J.L., J.H. Sock and.m. Soker (989), Seiparaeric esiaion of index coefficiens, Econoerica 57, Robinson, P. (989), Hypohesis esing in seiparaeric and nonparaeric odels for econoeric ie series, Review of Econoic Sudies 56, Sheran, R. (994), Maxial inequaliies for degenerae U-processes wih applicaions o opiizaion esiaors, Annals of Saisics, Zheng, J. (998), A consisen nonparaeric es of paraeric regression odels under condiional quanile resricions, Econoeric heory 4,

26 SFB 649 Discussion Paper Series 8 For a coplee lis of Discussion Papers published by he SFB 649, please visi hp://sfb649.wiwi.hu-berlin.de. "esing Monooniciy of Pricing Kernels" by Yuri Golubev, Wolfgang Härdle and Roan ionfeev, January 8. "Adapive poinwise esiaion in ie-inhoogeneous ie-series odels" by Pavel Cizek, Wolfgang Härdle and Vladiir Spokoiny, January 8. 3 "he Bayesian Addiive Classificaion ree Applied o Credi Risk Modelling" by Junni L. Zhang and Wolfgang Härdle, January 8. 4 "Independen Coponen Analysis Via Copula echniques" by Ray-Bing Chen, Meihui Guo, Wolfgang Härdle and Shih-Feng Huang, January 8. 5 "he Defaul Risk of Firs Exained wih Sooh Suppor Vecor Machines" by Wolfgang Härdle, Yuh-Jye Lee, Dorohea Schäfer and Yi-Ren Yeh, January 8. 6 "Value-a-Risk and Expeced Shorfall when here is long range dependence" by Wolfgang Härdle and Julius Mungo, Januray 8. 7 "A Consisen Nonparaeric es for Causaliy in Quanile" by Kiho Jeong and Wolfgang Härdle, January 8. SFB 649, Spandauer Sraße, D-78 Berlin hp://sfb649.wiwi.hu-berlin.de his research was suppored by he Deusche Forschungsgeeinschaf hrough he SFB 649 "Econoic Risk".

Lecture 18 GMM:IV, Nonlinear Models

Lecture 18 GMM:IV, Nonlinear Models Lecure 8 :IV, Nonlinear Models Le Z, be an rx funcion of a kx paraeer vecor, r > k, and a rando vecor Z, such ha he r populaion oen condiions also called esiain equaions EZ, hold for all, where is he rue

More information

Problem set 2 for the course on. Markov chains and mixing times

Problem set 2 for the course on. Markov chains and mixing times J. Seif T. Hirscher Soluions o Proble se for he course on Markov chains and ixing ies February 7, 04 Exercise 7 (Reversible chains). (i) Assue ha we have a Markov chain wih ransiion arix P, such ha here

More information

THE FINITE HAUSDORFF AND FRACTAL DIMENSIONS OF THE GLOBAL ATTRACTOR FOR A CLASS KIRCHHOFF-TYPE EQUATIONS

THE FINITE HAUSDORFF AND FRACTAL DIMENSIONS OF THE GLOBAL ATTRACTOR FOR A CLASS KIRCHHOFF-TYPE EQUATIONS European Journal of Maheaics and Copuer Science Vol 4 No 7 ISSN 59-995 HE FINIE HAUSDORFF AND FRACAL DIMENSIONS OF HE GLOBAL ARACOR FOR A CLASS KIRCHHOFF-YPE EQUAIONS Guoguang Lin & Xiangshuang Xia Deparen

More information

Estimates and Forecasts of GARCH Model under Misspecified Probability Distributions: A Monte Carlo Simulation Approach

Estimates and Forecasts of GARCH Model under Misspecified Probability Distributions: A Monte Carlo Simulation Approach Journal of Modern Applied Saisical Mehods Volue 3 Issue Aricle 8-04 Esiaes and Forecass of GARCH Model under Misspecified Probabiliy Disribuions: A Mone Carlo Siulaion Approach OlaOluwa S. Yaya Universiy

More information

Last exit times for a class of asymptotically linear estimators

Last exit times for a class of asymptotically linear estimators Las exi ies for a class of asypoically linear esiaors M. Alagh, M. Broniaowski 2, and G. Celan 3 Dépareen de Mahéaiques, Universié Louis Paseur, 4 Rue René Descares, 67000 Srasbourg, France 2 LSTA, Universié

More information

Testing the Random Walk Model. i.i.d. ( ) r

Testing the Random Walk Model. i.i.d. ( ) r he random walk heory saes: esing he Random Walk Model µ ε () np = + np + Momen Condiions where where ε ~ i.i.d he idea here is o es direcly he resricions imposed by momen condiions. lnp lnp µ ( lnp lnp

More information

1 Widrow-Hoff Algorithm

1 Widrow-Hoff Algorithm COS 511: heoreical Machine Learning Lecurer: Rob Schapire Lecure # 18 Scribe: Shaoqing Yang April 10, 014 1 Widrow-Hoff Algorih Firs le s review he Widrow-Hoff algorih ha was covered fro las lecure: Algorih

More information

Testing for a Single Factor Model in the Multivariate State Space Framework

Testing for a Single Factor Model in the Multivariate State Space Framework esing for a Single Facor Model in he Mulivariae Sae Space Framework Chen C.-Y. M. Chiba and M. Kobayashi Inernaional Graduae School of Social Sciences Yokohama Naional Universiy Japan Faculy of Economics

More information

A Generalization of Student s t-distribution from the Viewpoint of Special Functions

A Generalization of Student s t-distribution from the Viewpoint of Special Functions A Generalizaion of Suden s -disribuion fro he Viewpoin of Special Funcions WOLFRAM KOEPF and MOHAMMAD MASJED-JAMEI Deparen of Maheaics, Universiy of Kassel, Heinrich-Ple-Sr. 4, D-343 Kassel, Gerany Deparen

More information

Fourier Series & The Fourier Transform. Joseph Fourier, our hero. Lord Kelvin on Fourier s theorem. What do we want from the Fourier Transform?

Fourier Series & The Fourier Transform. Joseph Fourier, our hero. Lord Kelvin on Fourier s theorem. What do we want from the Fourier Transform? ourier Series & The ourier Transfor Wha is he ourier Transfor? Wha do we wan fro he ourier Transfor? We desire a easure of he frequencies presen in a wave. This will lead o a definiion of he er, he specru.

More information

Introduction to Numerical Analysis. In this lesson you will be taken through a pair of techniques that will be used to solve the equations of.

Introduction to Numerical Analysis. In this lesson you will be taken through a pair of techniques that will be used to solve the equations of. Inroducion o Nuerical Analysis oion In his lesson you will be aen hrough a pair of echniques ha will be used o solve he equaions of and v dx d a F d for siuaions in which F is well nown, and he iniial

More information

b denotes trend at time point t and it is sum of two

b denotes trend at time point t and it is sum of two Inernaional Conference on Innovaive Applicaions in Engineering and Inforaion echnology(iciaei207) Inernaional Journal of Advanced Scienific echnologies,engineering and Manageen Sciences (IJASEMSISSN: 2454356X)

More information

An Introduction to Malliavin calculus and its applications

An Introduction to Malliavin calculus and its applications An Inroducion o Malliavin calculus and is applicaions Lecure 5: Smoohness of he densiy and Hörmander s heorem David Nualar Deparmen of Mahemaics Kansas Universiy Universiy of Wyoming Summer School 214

More information

On Edgeworth Expansions in Generalized Urn Models

On Edgeworth Expansions in Generalized Urn Models On Edgeworh Expansions in Generalized Urn Models Sh M Mirahedov Insiue of Maheaics and Inforaion Technologies Uzbeisan (E- ail: shirahedov@yahooco S Rao Jaalaadaa Universiy of California Sana Barbara USA

More information

The Asymptotic Behavior of Nonoscillatory Solutions of Some Nonlinear Dynamic Equations on Time Scales

The Asymptotic Behavior of Nonoscillatory Solutions of Some Nonlinear Dynamic Equations on Time Scales Advances in Dynamical Sysems and Applicaions. ISSN 0973-5321 Volume 1 Number 1 (2006, pp. 103 112 c Research India Publicaions hp://www.ripublicaion.com/adsa.hm The Asympoic Behavior of Nonoscillaory Soluions

More information

arxiv: v1 [math.fa] 12 Jul 2012

arxiv: v1 [math.fa] 12 Jul 2012 AN EXTENSION OF THE LÖWNER HEINZ INEQUALITY MOHAMMAD SAL MOSLEHIAN AND HAMED NAJAFI arxiv:27.2864v [ah.fa] 2 Jul 22 Absrac. We exend he celebraed Löwner Heinz inequaliy by showing ha if A, B are Hilber

More information

Empirical Process Theory

Empirical Process Theory Empirical Process heory 4.384 ime Series Analysis, Fall 27 Reciaion by Paul Schrimpf Supplemenary o lecures given by Anna Mikusheva Ocober 7, 28 Reciaion 7 Empirical Process heory Le x be a real-valued

More information

Outline. lse-logo. Outline. Outline. 1 Wald Test. 2 The Likelihood Ratio Test. 3 Lagrange Multiplier Tests

Outline. lse-logo. Outline. Outline. 1 Wald Test. 2 The Likelihood Ratio Test. 3 Lagrange Multiplier Tests Ouline Ouline Hypohesis Tes wihin he Maximum Likelihood Framework There are hree main frequenis approaches o inference wihin he Maximum Likelihood framework: he Wald es, he Likelihood Raio es and he Lagrange

More information

Unit Root Time Series. Univariate random walk

Unit Root Time Series. Univariate random walk Uni Roo ime Series Univariae random walk Consider he regression y y where ~ iid N 0, he leas squares esimae of is: ˆ yy y y yy Now wha if = If y y hen le y 0 =0 so ha y j j If ~ iid N 0, hen y ~ N 0, he

More information

A Specification Test for Linear Dynamic Stochastic General Equilibrium Models

A Specification Test for Linear Dynamic Stochastic General Equilibrium Models Journal of Saisical and Economeric Mehods, vol.1, no.2, 2012, 65-70 ISSN: 2241-0384 (prin), 2241-0376 (online) Scienpress Ld, 2012 A Specificaion Tes for Linear Dynamic Sochasic General Equilibrium Models

More information

MODULE 3 FUNCTION OF A RANDOM VARIABLE AND ITS DISTRIBUTION LECTURES PROBABILITY DISTRIBUTION OF A FUNCTION OF A RANDOM VARIABLE

MODULE 3 FUNCTION OF A RANDOM VARIABLE AND ITS DISTRIBUTION LECTURES PROBABILITY DISTRIBUTION OF A FUNCTION OF A RANDOM VARIABLE Topics MODULE 3 FUNCTION OF A RANDOM VARIABLE AND ITS DISTRIBUTION LECTURES 2-6 3. FUNCTION OF A RANDOM VARIABLE 3.2 PROBABILITY DISTRIBUTION OF A FUNCTION OF A RANDOM VARIABLE 3.3 EXPECTATION AND MOMENTS

More information

Ann. Funct. Anal. 2 (2011), no. 2, A nnals of F unctional A nalysis ISSN: (electronic) URL:

Ann. Funct. Anal. 2 (2011), no. 2, A nnals of F unctional A nalysis ISSN: (electronic) URL: Ann. Func. Anal. 2 2011, no. 2, 34 41 A nnals of F uncional A nalysis ISSN: 2008-8752 elecronic URL: www.emis.de/journals/afa/ CLASSIFICAION OF POSIIVE SOLUIONS OF NONLINEAR SYSEMS OF VOLERRA INEGRAL EQUAIONS

More information

Utility maximization in incomplete markets

Utility maximization in incomplete markets Uiliy maximizaion in incomplee markes Marcel Ladkau 27.1.29 Conens 1 Inroducion and general seings 2 1.1 Marke model....................................... 2 1.2 Trading sraegy.....................................

More information

A revisit on the role of macro imbalances in the US recession of Aviral Kumar Tiwari # ICFAI University Tripura, India

A revisit on the role of macro imbalances in the US recession of Aviral Kumar Tiwari # ICFAI University Tripura, India A revisi on he role of acro ibalances in he US recession of 2007-2009 Aviral Kuar Tiwari # ICFAI Universiy Tripura, India Absrac The presen sudy is an aep o revisi he evidences of a very recen sudy of

More information

On the approximation of particular solution of nonhomogeneous linear differential equation with Legendre series

On the approximation of particular solution of nonhomogeneous linear differential equation with Legendre series The Journal of Applied Science Vol. 5 No. : -9 [6] วารสารว ทยาศาสตร ประย กต doi:.446/j.appsci.6.8. ISSN 53-785 Prined in Thailand Research Aricle On he approxiaion of paricular soluion of nonhoogeneous

More information

arxiv: v1 [math.fa] 9 Dec 2018

arxiv: v1 [math.fa] 9 Dec 2018 AN INVERSE FUNCTION THEOREM CONVERSE arxiv:1812.03561v1 [mah.fa] 9 Dec 2018 JIMMIE LAWSON Absrac. We esablish he following converse of he well-known inverse funcion heorem. Le g : U V and f : V U be inverse

More information

R t. C t P t. + u t. C t = αp t + βr t + v t. + β + w t

R t. C t P t. + u t. C t = αp t + βr t + v t. + β + w t Exercise 7 C P = α + β R P + u C = αp + βr + v (a) (b) C R = α P R + β + w (c) Assumpions abou he disurbances u, v, w : Classical assumions on he disurbance of one of he equaions, eg. on (b): E(v v s P,

More information

t 2 B F x,t n dsdt t u x,t dxdt

t 2 B F x,t n dsdt t u x,t dxdt Evoluion Equaions For 0, fixed, le U U0, where U denoes a bounded open se in R n.suppose ha U is filled wih a maerial in which a conaminan is being ranspored by various means including diffusion and convecion.

More information

Vehicle Arrival Models : Headway

Vehicle Arrival Models : Headway Chaper 12 Vehicle Arrival Models : Headway 12.1 Inroducion Modelling arrival of vehicle a secion of road is an imporan sep in raffic flow modelling. I has imporan applicaion in raffic flow simulaion where

More information

arxiv:math/ v1 [math.nt] 3 Nov 2005

arxiv:math/ v1 [math.nt] 3 Nov 2005 arxiv:mah/0511092v1 [mah.nt] 3 Nov 2005 A NOTE ON S AND THE ZEROS OF THE RIEMANN ZETA-FUNCTION D. A. GOLDSTON AND S. M. GONEK Absrac. Le πs denoe he argumen of he Riemann zea-funcion a he poin 1 + i. Assuming

More information

How to Deal with Structural Breaks in Practical Cointegration Analysis

How to Deal with Structural Breaks in Practical Cointegration Analysis How o Deal wih Srucural Breaks in Pracical Coinegraion Analysis Roselyne Joyeux * School of Economic and Financial Sudies Macquarie Universiy December 00 ABSTRACT In his noe we consider he reamen of srucural

More information

TIME DELAY BASEDUNKNOWN INPUT OBSERVER DESIGN FOR NETWORK CONTROL SYSTEM

TIME DELAY BASEDUNKNOWN INPUT OBSERVER DESIGN FOR NETWORK CONTROL SYSTEM TIME DELAY ASEDUNKNOWN INPUT OSERVER DESIGN FOR NETWORK CONTROL SYSTEM Siddhan Chopra J.S. Laher Elecrical Engineering Deparen NIT Kurukshera (India Elecrical Engineering Deparen NIT Kurukshera (India

More information

Vectorautoregressive Model and Cointegration Analysis. Time Series Analysis Dr. Sevtap Kestel 1

Vectorautoregressive Model and Cointegration Analysis. Time Series Analysis Dr. Sevtap Kestel 1 Vecorauoregressive Model and Coinegraion Analysis Par V Time Series Analysis Dr. Sevap Kesel 1 Vecorauoregression Vecor auoregression (VAR) is an economeric model used o capure he evoluion and he inerdependencies

More information

Econ107 Applied Econometrics Topic 7: Multicollinearity (Studenmund, Chapter 8)

Econ107 Applied Econometrics Topic 7: Multicollinearity (Studenmund, Chapter 8) I. Definiions and Problems A. Perfec Mulicollineariy Econ7 Applied Economerics Topic 7: Mulicollineariy (Sudenmund, Chaper 8) Definiion: Perfec mulicollineariy exiss in a following K-variable regression

More information

Richard A. Davis Colorado State University Bojan Basrak Eurandom Thomas Mikosch University of Groningen

Richard A. Davis Colorado State University Bojan Basrak Eurandom Thomas Mikosch University of Groningen Mulivariae Regular Variaion wih Applicaion o Financial Time Series Models Richard A. Davis Colorado Sae Universiy Bojan Basrak Eurandom Thomas Mikosch Universiy of Groningen Ouline + Characerisics of some

More information

Oscillation Properties of a Logistic Equation with Several Delays

Oscillation Properties of a Logistic Equation with Several Delays Journal of Maheaical Analysis and Applicaions 247, 11 125 Ž 2. doi:1.16 jaa.2.683, available online a hp: www.idealibrary.co on Oscillaion Properies of a Logisic Equaion wih Several Delays Leonid Berezansy

More information

T L. t=1. Proof of Lemma 1. Using the marginal cost accounting in Equation(4) and standard arguments. t )+Π RB. t )+K 1(Q RB

T L. t=1. Proof of Lemma 1. Using the marginal cost accounting in Equation(4) and standard arguments. t )+Π RB. t )+K 1(Q RB Elecronic Companion EC.1. Proofs of Technical Lemmas and Theorems LEMMA 1. Le C(RB) be he oal cos incurred by he RB policy. Then we have, T L E[C(RB)] 3 E[Z RB ]. (EC.1) Proof of Lemma 1. Using he marginal

More information

Granger Causality Among Pre-Crisis East Asian Exchange Rates. (Running Title: Granger Causality Among Pre-Crisis East Asian Exchange Rates)

Granger Causality Among Pre-Crisis East Asian Exchange Rates. (Running Title: Granger Causality Among Pre-Crisis East Asian Exchange Rates) Granger Causaliy Among PreCrisis Eas Asian Exchange Raes (Running Tile: Granger Causaliy Among PreCrisis Eas Asian Exchange Raes) Joseph D. ALBA and Donghyun PARK *, School of Humaniies and Social Sciences

More information

OBJECTIVES OF TIME SERIES ANALYSIS

OBJECTIVES OF TIME SERIES ANALYSIS OBJECTIVES OF TIME SERIES ANALYSIS Undersanding he dynamic or imedependen srucure of he observaions of a single series (univariae analysis) Forecasing of fuure observaions Asceraining he leading, lagging

More information

Heavy Tails of Discounted Aggregate Claims in the Continuous-time Renewal Model

Heavy Tails of Discounted Aggregate Claims in the Continuous-time Renewal Model Heavy Tails of Discouned Aggregae Claims in he Coninuous-ime Renewal Model Qihe Tang Deparmen of Saisics and Acuarial Science The Universiy of Iowa 24 Schae er Hall, Iowa Ciy, IA 52242, USA E-mail: qang@sa.uiowa.edu

More information

Regression with Time Series Data

Regression with Time Series Data Regression wih Time Series Daa y = β 0 + β 1 x 1 +...+ β k x k + u Serial Correlaion and Heeroskedasiciy Time Series - Serial Correlaion and Heeroskedasiciy 1 Serially Correlaed Errors: Consequences Wih

More information

DEPARTMENT OF ECONOMICS AND FINANCE COLLEGE OF BUSINESS AND ECONOMICS UNIVERSITY OF CANTERBURY CHRISTCHURCH, NEW ZEALAND

DEPARTMENT OF ECONOMICS AND FINANCE COLLEGE OF BUSINESS AND ECONOMICS UNIVERSITY OF CANTERBURY CHRISTCHURCH, NEW ZEALAND DEPARTMENT OF ECONOMICS AND FINANCE COLLEGE OF BUSINESS AND ECONOMICS UNIVERSITY OF CANTERBURY CHRISTCHURCH, NEW ZEALAND Asymmery and Leverage in Condiional Volailiy Models Michael McAleer WORKING PAPER

More information

International Journal of Scientific & Engineering Research, Volume 4, Issue 10, October ISSN

International Journal of Scientific & Engineering Research, Volume 4, Issue 10, October ISSN Inernaional Journal of Scienific & Engineering Research, Volume 4, Issue 10, Ocober-2013 900 FUZZY MEAN RESIDUAL LIFE ORDERING OF FUZZY RANDOM VARIABLES J. EARNEST LAZARUS PIRIYAKUMAR 1, A. YAMUNA 2 1.

More information

Comparing Means: t-tests for One Sample & Two Related Samples

Comparing Means: t-tests for One Sample & Two Related Samples Comparing Means: -Tess for One Sample & Two Relaed Samples Using he z-tes: Assumpions -Tess for One Sample & Two Relaed Samples The z-es (of a sample mean agains a populaion mean) is based on he assumpion

More information

Chapter 2. First Order Scalar Equations

Chapter 2. First Order Scalar Equations Chaper. Firs Order Scalar Equaions We sar our sudy of differenial equaions in he same way he pioneers in his field did. We show paricular echniques o solve paricular ypes of firs order differenial equaions.

More information

arxiv: v1 [math.pr] 19 Feb 2011

arxiv: v1 [math.pr] 19 Feb 2011 A NOTE ON FELLER SEMIGROUPS AND RESOLVENTS VADIM KOSTRYKIN, JÜRGEN POTTHOFF, AND ROBERT SCHRADER ABSTRACT. Various equivalen condiions for a semigroup or a resolven generaed by a Markov process o be of

More information

On a Fractional Stochastic Landau-Ginzburg Equation

On a Fractional Stochastic Landau-Ginzburg Equation Applied Mahemaical Sciences, Vol. 4, 1, no. 7, 317-35 On a Fracional Sochasic Landau-Ginzburg Equaion Nguyen Tien Dung Deparmen of Mahemaics, FPT Universiy 15B Pham Hung Sree, Hanoi, Vienam dungn@fp.edu.vn

More information

Convergence of the Neumann series in higher norms

Convergence of the Neumann series in higher norms Convergence of he Neumann series in higher norms Charles L. Epsein Deparmen of Mahemaics, Universiy of Pennsylvania Version 1.0 Augus 1, 003 Absrac Naural condiions on an operaor A are given so ha he Neumann

More information

EXERCISES FOR SECTION 1.5

EXERCISES FOR SECTION 1.5 1.5 Exisence and Uniqueness of Soluions 43 20. 1 v c 21. 1 v c 1 2 4 6 8 10 1 2 2 4 6 8 10 Graph of approximae soluion obained using Euler s mehod wih = 0.1. Graph of approximae soluion obained using Euler

More information

The General Linear Test in the Ridge Regression

The General Linear Test in the Ridge Regression ommunicaions for Saisical Applicaions Mehods 2014, Vol. 21, No. 4, 297 307 DOI: hp://dx.doi.org/10.5351/sam.2014.21.4.297 Prin ISSN 2287-7843 / Online ISSN 2383-4757 The General Linear Tes in he Ridge

More information

2.1 Level, Weight, Nominator and Denominator of an Eta Product. By an eta product we understand any finite product of functions. f(z) = m.

2.1 Level, Weight, Nominator and Denominator of an Eta Product. By an eta product we understand any finite product of functions. f(z) = m. Ea Producs.1 Level, Weigh, Noinaor and Denoinaor of an Ea Produc By an ea produc we undersand any finie produc of funcions f(z = η(z a where runs hrough a finie se of posiive inegers and he exponens a

More information

23.5. Half-Range Series. Introduction. Prerequisites. Learning Outcomes

23.5. Half-Range Series. Introduction. Prerequisites. Learning Outcomes Half-Range Series 2.5 Inroducion In his Secion we address he following problem: Can we find a Fourier series expansion of a funcion defined over a finie inerval? Of course we recognise ha such a funcion

More information

Matrix Versions of Some Refinements of the Arithmetic-Geometric Mean Inequality

Matrix Versions of Some Refinements of the Arithmetic-Geometric Mean Inequality Marix Versions of Some Refinemens of he Arihmeic-Geomeric Mean Inequaliy Bao Qi Feng and Andrew Tonge Absrac. We esablish marix versions of refinemens due o Alzer ], Carwrigh and Field 4], and Mercer 5]

More information

23.2. Representing Periodic Functions by Fourier Series. Introduction. Prerequisites. Learning Outcomes

23.2. Representing Periodic Functions by Fourier Series. Introduction. Prerequisites. Learning Outcomes Represening Periodic Funcions by Fourier Series 3. Inroducion In his Secion we show how a periodic funcion can be expressed as a series of sines and cosines. We begin by obaining some sandard inegrals

More information

Robust estimation based on the first- and third-moment restrictions of the power transformation model

Robust estimation based on the first- and third-moment restrictions of the power transformation model h Inernaional Congress on Modelling and Simulaion, Adelaide, Ausralia, 6 December 3 www.mssanz.org.au/modsim3 Robus esimaion based on he firs- and hird-momen resricions of he power ransformaion Nawaa,

More information

POSITIVE SOLUTIONS OF NEUTRAL DELAY DIFFERENTIAL EQUATION

POSITIVE SOLUTIONS OF NEUTRAL DELAY DIFFERENTIAL EQUATION Novi Sad J. Mah. Vol. 32, No. 2, 2002, 95-108 95 POSITIVE SOLUTIONS OF NEUTRAL DELAY DIFFERENTIAL EQUATION Hajnalka Péics 1, János Karsai 2 Absrac. We consider he scalar nonauonomous neural delay differenial

More information

Stability and Bifurcation in a Neural Network Model with Two Delays

Stability and Bifurcation in a Neural Network Model with Two Delays Inernaional Mahemaical Forum, Vol. 6, 11, no. 35, 175-1731 Sabiliy and Bifurcaion in a Neural Nework Model wih Two Delays GuangPing Hu and XiaoLing Li School of Mahemaics and Physics, Nanjing Universiy

More information

Wave Mechanics. January 16, 2017

Wave Mechanics. January 16, 2017 Wave Mechanics January 6, 7 The ie-dependen Schrödinger equaion We have seen how he ie-dependen Schrodinger equaion, Ψ + Ψ i Ψ follows as a non-relaivisic version of he Klein-Gordon equaion. In wave echanics,

More information

A note on spurious regressions between stationary series

A note on spurious regressions between stationary series A noe on spurious regressions beween saionary series Auhor Su, Jen-Je Published 008 Journal Tile Applied Economics Leers DOI hps://doi.org/10.1080/13504850601018106 Copyrigh Saemen 008 Rouledge. This is

More information

Note on oscillation conditions for first-order delay differential equations

Note on oscillation conditions for first-order delay differential equations Elecronic Journal of Qualiaive Theory of Differenial Equaions 2016, No. 2, 1 10; doi: 10.14232/ejqde.2016.1.2 hp://www.ah.u-szeged.hu/ejqde/ Noe on oscillaion condiions for firs-order delay differenial

More information

Approximate Message Passing with Consistent Parameter Estimation and Applications to Sparse Learning

Approximate Message Passing with Consistent Parameter Estimation and Applications to Sparse Learning APPROXIMATE MESSAGE PASSING WITH CONSISTENT PARAMETER ESTIMATION Approxiae Message Passing wih Consisen Paraeer Esiaion and Applicaions o Sparse Learning Ulugbe S. Kailov, Suden Meber, IEEE, Sundeep Rangan,

More information

ACE 564 Spring Lecture 7. Extensions of The Multiple Regression Model: Dummy Independent Variables. by Professor Scott H.

ACE 564 Spring Lecture 7. Extensions of The Multiple Regression Model: Dummy Independent Variables. by Professor Scott H. ACE 564 Spring 2006 Lecure 7 Exensions of The Muliple Regression Model: Dumm Independen Variables b Professor Sco H. Irwin Readings: Griffihs, Hill and Judge. "Dumm Variables and Varing Coefficien Models

More information

Hypothesis Testing in the Classical Normal Linear Regression Model. 1. Components of Hypothesis Tests

Hypothesis Testing in the Classical Normal Linear Regression Model. 1. Components of Hypothesis Tests ECONOMICS 35* -- NOTE 8 M.G. Abbo ECON 35* -- NOTE 8 Hypohesis Tesing in he Classical Normal Linear Regression Model. Componens of Hypohesis Tess. A esable hypohesis, which consiss of wo pars: Par : a

More information

Stochastic Model for Cancer Cell Growth through Single Forward Mutation

Stochastic Model for Cancer Cell Growth through Single Forward Mutation Journal of Modern Applied Saisical Mehods Volume 16 Issue 1 Aricle 31 5-1-2017 Sochasic Model for Cancer Cell Growh hrough Single Forward Muaion Jayabharahiraj Jayabalan Pondicherry Universiy, jayabharahi8@gmail.com

More information

Properties Of Solutions To A Generalized Liénard Equation With Forcing Term

Properties Of Solutions To A Generalized Liénard Equation With Forcing Term Applied Mahemaics E-Noes, 8(28), 4-44 c ISSN 67-25 Available free a mirror sies of hp://www.mah.nhu.edu.w/ amen/ Properies Of Soluions To A Generalized Liénard Equaion Wih Forcing Term Allan Kroopnick

More information

Finish reading Chapter 2 of Spivak, rereading earlier sections as necessary. handout and fill in some missing details!

Finish reading Chapter 2 of Spivak, rereading earlier sections as necessary. handout and fill in some missing details! MAT 257, Handou 6: Ocober 7-2, 20. I. Assignmen. Finish reading Chaper 2 of Spiva, rereading earlier secions as necessary. handou and fill in some missing deails! II. Higher derivaives. Also, read his

More information

arxiv: v1 [math.pr] 23 Jan 2019

arxiv: v1 [math.pr] 23 Jan 2019 Consrucion of Liouville Brownian moion via Dirichle form heory Jiyong Shin arxiv:90.07753v [mah.pr] 23 Jan 209 Absrac. The Liouville Brownian moion which was inroduced in [3] is a naural diffusion process

More information

LECTURE 1: GENERALIZED RAY KNIGHT THEOREM FOR FINITE MARKOV CHAINS

LECTURE 1: GENERALIZED RAY KNIGHT THEOREM FOR FINITE MARKOV CHAINS LECTURE : GENERALIZED RAY KNIGHT THEOREM FOR FINITE MARKOV CHAINS We will work wih a coninuous ime reversible Markov chain X on a finie conneced sae space, wih generaor Lf(x = y q x,yf(y. (Recall ha q

More information

On Gronwall s Type Integral Inequalities with Singular Kernels

On Gronwall s Type Integral Inequalities with Singular Kernels Filoma 31:4 (217), 141 149 DOI 1.2298/FIL17441A Published by Faculy of Sciences and Mahemaics, Universiy of Niš, Serbia Available a: hp://www.pmf.ni.ac.rs/filoma On Gronwall s Type Inegral Inequaliies

More information

Math 10B: Mock Mid II. April 13, 2016

Math 10B: Mock Mid II. April 13, 2016 Name: Soluions Mah 10B: Mock Mid II April 13, 016 1. ( poins) Sae, wih jusificaion, wheher he following saemens are rue or false. (a) If a 3 3 marix A saisfies A 3 A = 0, hen i canno be inverible. True.

More information

Differential Harnack Estimates for Parabolic Equations

Differential Harnack Estimates for Parabolic Equations Differenial Harnack Esimaes for Parabolic Equaions Xiaodong Cao and Zhou Zhang Absrac Le M,g be a soluion o he Ricci flow on a closed Riemannian manifold In his paper, we prove differenial Harnack inequaliies

More information

Homework 2 Solutions

Homework 2 Solutions Mah 308 Differenial Equaions Fall 2002 & 2. See he las page. Hoework 2 Soluions 3a). Newon s secon law of oion says ha a = F, an we know a =, so we have = F. One par of he force is graviy, g. However,

More information

I-Optimal designs for third degree kronecker model mixture experiments

I-Optimal designs for third degree kronecker model mixture experiments Inernaional Journal of Saisics and Applied Maheaics 207 2(2): 5-40 ISSN: 2456-452 Mahs 207 2(2): 5-40 207 Sas & Mahs www.ahsjournal.co Received: 9-0-207 Acceped: 20-02-207 Cheruiyo Kipkoech Deparen of

More information

BOUNDEDNESS OF MAXIMAL FUNCTIONS ON NON-DOUBLING MANIFOLDS WITH ENDS

BOUNDEDNESS OF MAXIMAL FUNCTIONS ON NON-DOUBLING MANIFOLDS WITH ENDS BOUNDEDNESS OF MAXIMAL FUNCTIONS ON NON-DOUBLING MANIFOLDS WITH ENDS XUAN THINH DUONG, JI LI, AND ADAM SIKORA Absrac Le M be a manifold wih ends consruced in [2] and be he Laplace-Belrami operaor on M

More information

Oscillation of an Euler Cauchy Dynamic Equation S. Huff, G. Olumolode, N. Pennington, and A. Peterson

Oscillation of an Euler Cauchy Dynamic Equation S. Huff, G. Olumolode, N. Pennington, and A. Peterson PROCEEDINGS OF THE FOURTH INTERNATIONAL CONFERENCE ON DYNAMICAL SYSTEMS AND DIFFERENTIAL EQUATIONS May 4 7, 00, Wilmingon, NC, USA pp 0 Oscillaion of an Euler Cauchy Dynamic Equaion S Huff, G Olumolode,

More information

Sample Autocorrelations for Financial Time Series Models. Richard A. Davis Colorado State University Thomas Mikosch University of Copenhagen

Sample Autocorrelations for Financial Time Series Models. Richard A. Davis Colorado State University Thomas Mikosch University of Copenhagen Sample Auocorrelaions for Financial Time Series Models Richard A. Davis Colorado Sae Universiy Thomas Mikosch Universiy of Copenhagen Ouline Characerisics of some financial ime series IBM reurns NZ-USA

More information

Endpoint Strichartz estimates

Endpoint Strichartz estimates Endpoin Sricharz esimaes Markus Keel and Terence Tao (Amer. J. Mah. 10 (1998) 955 980) Presener : Nobu Kishimoo (Kyoo Universiy) 013 Paricipaing School in Analysis of PDE 013/8/6 30, Jeju 1 Absrac of he

More information

Existence Theory of Second Order Random Differential Equations

Existence Theory of Second Order Random Differential Equations Global Journal of Mahemaical Sciences: Theory and Pracical. ISSN 974-32 Volume 4, Number 3 (22), pp. 33-3 Inernaional Research Publicaion House hp://www.irphouse.com Exisence Theory of Second Order Random

More information

2. Nonlinear Conservation Law Equations

2. Nonlinear Conservation Law Equations . Nonlinear Conservaion Law Equaions One of he clear lessons learned over recen years in sudying nonlinear parial differenial equaions is ha i is generally no wise o ry o aack a general class of nonlinear

More information

Cash Flow Valuation Mode Lin Discrete Time

Cash Flow Valuation Mode Lin Discrete Time IOSR Journal of Mahemaics (IOSR-JM) e-issn: 2278-5728,p-ISSN: 2319-765X, 6, Issue 6 (May. - Jun. 2013), PP 35-41 Cash Flow Valuaion Mode Lin Discree Time Olayiwola. M. A. and Oni, N. O. Deparmen of Mahemaics

More information

MATH 4330/5330, Fourier Analysis Section 6, Proof of Fourier s Theorem for Pointwise Convergence

MATH 4330/5330, Fourier Analysis Section 6, Proof of Fourier s Theorem for Pointwise Convergence MATH 433/533, Fourier Analysis Secion 6, Proof of Fourier s Theorem for Poinwise Convergence Firs, some commens abou inegraing periodic funcions. If g is a periodic funcion, g(x + ) g(x) for all real x,

More information

A Revisit on the Role of Macro Imbalances in. the US Recession of : Nonlinear Causality Approach

A Revisit on the Role of Macro Imbalances in. the US Recession of : Nonlinear Causality Approach Applied Maheaical Sciences, Vol. 7, 2013, no. 47, 2321-2330 HIKARI Ld, www.-hikari.co A Revisi on he Role of Macro Ibalances in he US Recession of 2007-2009: Nonlinear Causaliy Approach Aviral Kuar Tiwari

More information

10. State Space Methods

10. State Space Methods . Sae Space Mehods. Inroducion Sae space modelling was briefly inroduced in chaper. Here more coverage is provided of sae space mehods before some of heir uses in conrol sysem design are covered in he

More information

Financial Econometrics Jeffrey R. Russell Midterm Winter 2009 SOLUTIONS

Financial Econometrics Jeffrey R. Russell Midterm Winter 2009 SOLUTIONS Name SOLUTIONS Financial Economerics Jeffrey R. Russell Miderm Winer 009 SOLUTIONS You have 80 minues o complee he exam. Use can use a calculaor and noes. Try o fi all your work in he space provided. If

More information

A Nonexistence Result to a Cauchy Problem in Nonlinear One Dimensional Thermoelasticity

A Nonexistence Result to a Cauchy Problem in Nonlinear One Dimensional Thermoelasticity Journal of Maheaical Analysis and Applicaions 54, 7186 1 doi:1.16jaa..73, available online a hp:www.idealibrary.co on A Nonexisence Resul o a Cauchy Proble in Nonlinear One Diensional Theroelasiciy Mokhar

More information

A New Unit Root Test against Asymmetric ESTAR Nonlinearity with Smooth Breaks

A New Unit Root Test against Asymmetric ESTAR Nonlinearity with Smooth Breaks Iran. Econ. Rev. Vol., No., 08. pp. 5-6 A New Uni Roo es agains Asymmeric ESAR Nonlineariy wih Smooh Breaks Omid Ranjbar*, sangyao Chang, Zahra (Mila) Elmi 3, Chien-Chiang Lee 4 Received: December 7, 06

More information

Sobolev-type Inequality for Spaces L p(x) (R N )

Sobolev-type Inequality for Spaces L p(x) (R N ) In. J. Conemp. Mah. Sciences, Vol. 2, 27, no. 9, 423-429 Sobolev-ype Inequaliy for Spaces L p(x ( R. Mashiyev and B. Çekiç Universiy of Dicle, Faculy of Sciences and Ars Deparmen of Mahemaics, 228-Diyarbakir,

More information

Riemann Hypothesis and Primorial Number. Choe Ryong Gil

Riemann Hypothesis and Primorial Number. Choe Ryong Gil Rieann Hyohesis Priorial Nuber Choe Ryong Gil Dearen of Maheaics Universiy of Sciences Gwahak- dong Unjong Disric Pyongyang DPRKorea Eail; ryonggilchoe@sar-conek Augus 8 5 Absrac; In his aer we consider

More information

Block Diagram of a DCS in 411

Block Diagram of a DCS in 411 Informaion source Forma A/D From oher sources Pulse modu. Muliplex Bandpass modu. X M h: channel impulse response m i g i s i Digial inpu Digial oupu iming and synchronizaion Digial baseband/ bandpass

More information

f(s)dw Solution 1. Approximate f by piece-wise constant left-continuous non-random functions f n such that (f(s) f n (s)) 2 ds 0.

f(s)dw Solution 1. Approximate f by piece-wise constant left-continuous non-random functions f n such that (f(s) f n (s)) 2 ds 0. Advanced Financial Models Example shee 3 - Michaelmas 217 Michael Tehranchi Problem 1. Le f : [, R be a coninuous (non-random funcion and W a Brownian moion, and le σ 2 = f(s 2 ds and assume σ 2

More information

Heat kernel and Harnack inequality on Riemannian manifolds

Heat kernel and Harnack inequality on Riemannian manifolds Hea kernel and Harnack inequaliy on Riemannian manifolds Alexander Grigor yan UHK 11/02/2014 onens 1 Laplace operaor and hea kernel 1 2 Uniform Faber-Krahn inequaliy 3 3 Gaussian upper bounds 4 4 ean-value

More information

Introduction D P. r = constant discount rate, g = Gordon Model (1962): constant dividend growth rate.

Introduction D P. r = constant discount rate, g = Gordon Model (1962): constant dividend growth rate. Inroducion Gordon Model (1962): D P = r g r = consan discoun rae, g = consan dividend growh rae. If raional expecaions of fuure discoun raes and dividend growh vary over ime, so should he D/P raio. Since

More information

Optimality Conditions for Unconstrained Problems

Optimality Conditions for Unconstrained Problems 62 CHAPTER 6 Opimaliy Condiions for Unconsrained Problems 1 Unconsrained Opimizaion 11 Exisence Consider he problem of minimizing he funcion f : R n R where f is coninuous on all of R n : P min f(x) x

More information

3.1.3 INTRODUCTION TO DYNAMIC OPTIMIZATION: DISCRETE TIME PROBLEMS. A. The Hamiltonian and First-Order Conditions in a Finite Time Horizon

3.1.3 INTRODUCTION TO DYNAMIC OPTIMIZATION: DISCRETE TIME PROBLEMS. A. The Hamiltonian and First-Order Conditions in a Finite Time Horizon 3..3 INRODUCION O DYNAMIC OPIMIZAION: DISCREE IME PROBLEMS A. he Hamilonian and Firs-Order Condiions in a Finie ime Horizon Define a new funcion, he Hamilonian funcion, H. H he change in he oal value of

More information

ON NONLINEAR CROSS-DIFFUSION SYSTEMS: AN OPTIMAL TRANSPORT APPROACH

ON NONLINEAR CROSS-DIFFUSION SYSTEMS: AN OPTIMAL TRANSPORT APPROACH ON NONLINEAR CROSS-DIFFUSION SYSTEMS: AN OPTIMAL TRANSPORT APPROACH INWON KIM AND ALPÁR RICHÁRD MÉSZÁROS Absrac. We sudy a nonlinear, degenerae cross-diffusion odel which involves wo densiies wih wo differen

More information

Second-Order Boundary Value Problems of Singular Type

Second-Order Boundary Value Problems of Singular Type JOURNAL OF MATEMATICAL ANALYSIS AND APPLICATIONS 226, 4443 998 ARTICLE NO. AY98688 Seond-Order Boundary Value Probles of Singular Type Ravi P. Agarwal Deparen of Maheais, Naional Uniersiy of Singapore,

More information

Lecture 33: November 29

Lecture 33: November 29 36-705: Inermediae Saisics Fall 2017 Lecurer: Siva Balakrishnan Lecure 33: November 29 Today we will coninue discussing he boosrap, and hen ry o undersand why i works in a simple case. In he las lecure

More information

A NOTE ON S(t) AND THE ZEROS OF THE RIEMANN ZETA-FUNCTION

A NOTE ON S(t) AND THE ZEROS OF THE RIEMANN ZETA-FUNCTION Bull. London Mah. Soc. 39 2007 482 486 C 2007 London Mahemaical Sociey doi:10.1112/blms/bdm032 A NOTE ON S AND THE ZEROS OF THE RIEMANN ZETA-FUNCTION D. A. GOLDSTON and S. M. GONEK Absrac Le πs denoe he

More information

ACE 562 Fall Lecture 4: Simple Linear Regression Model: Specification and Estimation. by Professor Scott H. Irwin

ACE 562 Fall Lecture 4: Simple Linear Regression Model: Specification and Estimation. by Professor Scott H. Irwin ACE 56 Fall 005 Lecure 4: Simple Linear Regression Model: Specificaion and Esimaion by Professor Sco H. Irwin Required Reading: Griffihs, Hill and Judge. "Simple Regression: Economic and Saisical Model

More information

Physics 127b: Statistical Mechanics. Fokker-Planck Equation. Time Evolution

Physics 127b: Statistical Mechanics. Fokker-Planck Equation. Time Evolution Physics 7b: Saisical Mechanics Fokker-Planck Equaion The Langevin equaion approach o he evoluion of he velociy disribuion for he Brownian paricle migh leave you uncomforable. A more formal reamen of his

More information