Estimation of Markov Regime-Switching Regression Models with Endogenous Switching

Size: px
Start display at page:

Download "Estimation of Markov Regime-Switching Regression Models with Endogenous Switching"

Transcription

1 Esimion of Mrkov Regime-Swiching Regression Models wih Endogenous Swiching Chng-Jin Kim Kore Universiy nd Universiy of Wshingon Jeremy Piger Federl Reserve Bnk of S. Louis Richrd Srz Universiy of Wshingon July 3 Absrc Following Hmilon (989), esimion of Mrkov regime-swiching regressions nerly lwys relies on he ssumpion h he len se vrible conrolling he regime chnge is exogenous. We relx his ssumpion nd develop wo echniques o esime Mrkov-swiching models wih endogenous swiching. The firs exends he endogenous swiching models in Mddl nd Nelson (975) o he Hmilon (989) Mrkov regime-swiching regression. The second is bsed on he inerpreion of he endogenous swiching regression s regression model wih endogenous unobserved dummy vribles. In his cse he regression cn be esimed using insrumenl vribles echniques. For boh echniques, idenificion is chieved when he rnsiion probbiliies of he regime-swiching process re influenced by observed exogenous vribles. However, even in he fixed rnsiion probbiliy cse, idenificion is chieved if he se process is serilly dependen nd lgged vlues of he se process re exogenous. This is rue even hough he lgged se is unobserved. The bis correcion echniques lso dmi srighforwrd ess for endogeneiy. Mone Crlo experimens confirm h he proposed bis correcion echniques perform quie well in prcice. We pply he procedure o he voliliy feedbck model of equiy reurns given in Turner, Srz nd Nelson (989). Keywords: Endogeneiy, Regime-Swiching, Insrumenl Vribles JEL Clssificion: C3, C, G Kim: Dep. of Economics, Kore Universiy, Anm-Dong, Seongbuk-ku, Seoul, 36-7, Kore, (cjkim@kore.c.kr); Piger: Reserch Deprmen, Federl Reserve Bnk of S. Louis, 4 Locus S., S. Louis, MO 6366, USA, (piger@sls.frb.org); Srz: Dep. of Economics, Universiy of Wshingon, Box 35333, Sele, WA 9895, USA (srz@u.wshingon.edu). We hnk Michel Dueker for helpful commens. The views expressed in his pper re solely he responsibiliy of he uhors, nd should no be inerpreed s reflecing he views of he Federl Reserve Bnk of S. Louis or of ny oher person ssocied wih he Federl Reserve Sysem.

2 There is mouning evidence h empiricl models of mny economic ime series, priculrly mcroeconomic nd finncil series, re chrcerized by prmeer insbiliy. This hs sprked n explosion of ineres in ime-vrying prmeer models. One noble se of models re swiching regressions wih unknown smple seprion, in which prmeers move discreely beween fixed number of regimes, wih he swiching conrolled by n unobserved se vrible. Swiching regressions hve rich hisory in economerics, ding bck o les Qund (958). Goldfeld nd Qund (973) inroduced priculrly useful version of hese models, referred o in he following s Mrkov-swiching model, in which he len se vrible conrolling he regime shifs follows Mrkov-chin, nd is hus serilly dependen. In n influenil ricle, Hmilon (989) exended Mrkov-swiching models o he cse of dependen d, specificlly n uoregression. The vs lierure spwned by Hmilon (989) hs ypiclly ssumed h he regime shifs re exogenous of ll relizions of he regression error. However, he erlier lierure on swiching regressions, such s Mddl nd Nelson (975), ws ofen concerned wih endogenous swiching, s he primry pplicions of swiching regression models in his lierure were in limied dependen vrible conexs such s self-selecion nd mrke disequilibrium problems. In his pper we work wih swiching regressions of he ype considered by Hmilon (989) nd vrious exensions, bu relx he exogenous swiching ssumpion. We show h filure of he exogeneiy ssumpion leds o significn bis in he coefficiens of regimeswiching regression when esimion mehods bsed on he exogeneiy ssumpion re used. We hen consider wo esimion echniques for correcing his bis. In he firs, we exend he endogenous swiching models in Mddl nd Nelson (975) o he Hmilon (989) regime

3 swiching regression, in which he swiching process is serilly dependen, specificlly Mrkovswiching. This pproch relies on join normliy ssumpion for he relionship beween he regression error erms nd he innovions o he se relizion equion. If one is unwilling o mke his ssumpion, we develop n lernive echnique, bsed on he inerpreion of he endogenous swiching regression s regression model wih endogenous unobserved dummy vribles. In his cse he regression cn be esimed using insrumenl vribles echniques. Boh bis correcion echniques lso dmi srighforwrd es for endogeneiy. For boh of hese esimion echniques, we show h for serilly dependen se processes, such s Mrkov-swiching se process, he lgged se cn provide informion necessry for idenificion, providing i is uncorreled wih he curren regression error. This is rue even hough he lgged se is unobserved. Addiionl informion is obined when he rnsiion probbiliies of he swiching process re influenced by exogenous vribles, s in he so clled ime-vrying rnsiion probbiliy cse. We use Mone Crlo experimens o confirm he efficcy of he proposed bis-correcion procedures nd find h heir performnce is quie good for empiriclly relevn d genering processes. Why re we moived o invesige Mrkov-swiching regressions wih endogenous swiching? Mny of he model s pplicions re in mcroeconomics or finnce in siuions where i would be nurl o ssume h he se is endogenous. As n exmple, in mny models he esimed se vrible hs srong business cycle correlion, ofen corresponding wih recessions. This cn be seen in recen pplicions of he regime-swiching model o idenified monery VARs, such s Sims nd Zh () nd Owyng (). I is no hrd o imgine h he shocks o he regression, such s he mcroeconomic shocks o he VAR, would be correled wih recessions. As noher exmple, some pplicions of he model conin

4 prmeers h represen he recion of gens o relizion of he se (see for exmple Turner, Srz nd Nelson (989)). However, i is likely h gens do no observe he se, bu insed drw inference bsed on some informion se, he conens of which re unknown o he economericin. Use of he cul se o proxy for his inference leds o regression wih mesuremen error in he explnory vribles, nd hus endogeneiy. In he nex secion we ly ou cnonicl Mrkov-swiching regression nd documen he bises h rise when he len se vrible is correled wih he regression error. Secion 3 develops he wo bis correcion sregies, discusses idenificion, nd presens ess for endogeneiy. Secion 4 gives he resuls of Mone Crlo experimen documening he performnce of he proposed bis correcion procedures. In Secion 5 we presen n empiricl exmple bsed on model of voliliy feedbck in equiy mrkes ken from Turner, Srz nd Nelson (989).. Endogenous Regime-Swiching nd Esimion Bis The model we consider is liner regression wih regime-swiching coefficiens nd residul vrince. Assume h he dependen vrible, y, is genered by one of wo rue regression equions: y = + x + ε, ε N(, σ ), ' S S S = ( S ) + S, S = ( S ) + S, S σ = σ ( S ) + σ S, S () We focus here on wo-regime model, however, ll of he resuls re esily generlized o he n regime cse. 3

5 where S = {, } is se vrible h indices which of he wo regression equions generes he observion y nd x is k x vecor of exogenous or predeermined explnory vribles mesured ime, which my include lgged vlues of y. We re ineresed in he cse of unobserved S nd hus require probbiliy lw governing S for esimion purposes. Here we consider regime-swiching process for S h encompsses severl populr specificions in he lierure. In priculr we ssume h he probbiliy h S = i depends on S nd on vecor of exogenous or predeermined vribles z, where z my include elemens of x. Formlly: PS ( = S =,..., S = iz, ) = PS ( = S =, z) = pz ( ) j PS ( = S =, z) = pz ( ) PS ( = S =,..., S = iz, ) = PS ( = S =, z) = qz ( ) () j PS ( = S =, z) = qz ( ) where he rnsiion probbiliies p z ) nd qz ( ) my hve eiher Probi or Logisic specificion. ( Mximum likelihood esimion of he model vi he recursive filer given in Hmilon (989) or he expecion mximizion (EM) lgorihm given in Hmilon (99) will yield consisen prmeer esimes under he ssumpion h S is uncorreled wih ε. Relxion of his ssumpion will led o bised esimion of he prmeers i nd, i =,. Noe i h, when ε is correled wih S, equion () cn be rewrien s follows: y = + x + E( ε S, S, z ) + u, ' S S u = ε E( ε S, S, z ) (3) 4

6 Th is, S inroduces predicbiliy ino ε h vries wih he regime swiching inercep. Thus, here is no informion o seprely idenify i from E( ε S = i, S, z). In generl, mximum likelihood esimion will yield bised esimes of boh i nd i. However, in he specil cse where x nd S re independen, he bis will be conined o i only. In he nex secion we presen echniques o correc he bis inroduced by he endogenous se vrible. 3. Bis Correcion in he Endogenous Swiching Model 3. Bis Correcion wih Join Normliy In his Secion, we prmeerize he rnsiion probbiliies p( z ) nd qz ( ) in equion () using he following probi specificion: S = if if S S < S = + S + z ' + z ' S + η (4) 3 η ~... iid N(,) So h: p( z ) = Pr( S = S =, z ) = Pr( η > ( ( + ) z '( + )) = Φ( ( + ) z '( + )) 3 3 qz ( ) = Pr( S= S =, z) = Pr( η z' ) =Φ( z' ) where Φ is he sndrd norml cumulive disribuion funcion. Severl specil cses of () re worh menioning. The unresriced model represens he ime-vrying rnsiion probbiliy Mrkov-swiching model (TVP-MS) of Goldfeld nd Qund (973), Diebold, Lee nd Weinbch (994) nd Filrdo (994). When = 3 =, we hve he fixed rnsiion probbiliy Mrkov-swiching model (FTP-MS) of Goldfeld nd Qund (973) nd Hmilon (989), h is 5

7 p( z ) = p, qz ( ) = q. If = = 3 = we hve he fixed rnsiion probbiliy independen swiching model (FTP-IS) of Qund (97), so h q= p. Finlly, if = 3 =, we hve he ime-vrying rnsiion probbiliy independen swiching model (TVP-IS) of Goldfeld nd Qund (97), q z ) = p( z ). ( Correlion beween ε nd S is inroduced hrough correlion beween ε nd η. Thus, he join disribuion beween ε nd η is mulivrie Norml wih he following covrince mrix: η ρσ S ~ N(, ), ε Σ Σ= ρσ σ S S (5) In his cse, he model in (), () nd (5) is closely reled o he swiching regression wih endogenous swiching considered by Mddl nd Nelson (975). The min ddiion we consider here is o llow he unobserved se process o be serilly dependen, specificlly Mrkov-swiching, s in Goldfeld nd Qund (973) nd Hmilon (989). In he following, we ouline mximum likelihood esimion procedure for his Mrkov-swiching regression wih endogenous swiching. Using he join norml ssumpion given in (5), one cn derive explici chrcerizions of he bis erms from equion (4), E( ε S = i, S, z). In priculr, noe from (5) h: 6

8 E( ε S =, S, z ) = E( ε η < S z ' z ' S ) 3 φ ( S z ' z ' S ) = ρσ = ρσ M 3 Φ ( S z ' z ' 3S ) E( ε S =, S, z) = E( ε η S z ' z ' 3S ) φ( S z ' z ' 3S ) = ρσ = ρσ M ( Φ( S z ' z ' S )) 3 (6) where φ is he sndrd norml densiy funcion. Using (6), we cn correc for bis inroduced by he endogenous se vrible by rewriing (3) s: y = + x + ρσ M + u, ' S S S S u = ε E( ε S, S, z ), (7) where M S, ( S =,), is defined in (6). The prmeers of (7) re only idenified if M nd M vry, oherwise here is no informion o idenify S from ρσ S M S, ( S, = ). From (6), we cn see his is chieved in wo wys. Firs, if he se process depends on exogenous or predeermined observble vribles z, he model is idenified. Second, even when = 3 =, s in he Mrkov-swiching model wih fixed rnsiion probbiliies, he model is sill idenified so long s. Th is, if S is serilly correled nd S is exogenous, S serves o idenify he model. Assuming idenificion, consisen esimes of he prmeers of () re chieved by mximum likelihood esimion of (7), which cn be performed compuionlly using he recursive filer given in Hmilon (989) or he expecion mximizion (EM) lgorihm discussed in Hmilon (99). As hese echniques re well known, we will no repe hem in deil here. Noe hough h compuion of he uncondiionl likelihood requires h he likelihood for (7) be compued condiionl on ech possible combinion of S nd S. This requires h we chrcerize he vrinces of u under ech se. These re given by: 7

9 vr( u S =, S, z ) = vr( ε S =, S, z ) = ρ σ ( M ( M + + S + z ' + z ' S )) + σ ( ρ ) 3 vr( u S =, S, z ) = vr( ε S =, S, z ) = ρ σ ( M ( M + + S + z ' + z ' S )) + σ ( ρ ) 3 (8) 3. Bis Correcion wihou Resoring o Join Normliy The bis correcion echnique presened in Secion 3. relied on specific disribuionl ssumpion describing he relionship beween ε nd η. In his secion, we presen n lernive bis correcion sregy, for which we do no resor o such disribuionl ssumpion. The procedure inroduced in his secion would work no only for he Probi specificion of he rnsiion probbiliies in equion (4), bu lso for he Logisic specificion of he rnsiion probbiliies of he following form: exp( + z ' b) pz ( ) = Pr( S = S =, z) = + exp( + z ' b) (9) exp( + z ' b ) qz = S= S = z = + + ( ) Pr(, ), exp( z ' b) where S is correled wih he disurbnce erm ε in equion (). In order o move owrd esimion wih bis correcion, we consider useful lernive AR() chrcerizion of S s suggesed by Hmilon (989): 8

10 S λ = c + S c = q( z ) λ = + p( z + v ) + q( z ) () where condiionl on S = : v = q( z )) wih probbiliy q z ) ( v = q z ) wih probbiliy q( z ) ( ( E( ν S = ) = Vr( ν S = ) = σ =, qz ( )( ν qz ( )) (.A) nd condiionl on S = : v = p( z )) wih probbiliy p z ) ( ( v = p z ) wih probbiliy p( z ) ( E( ν S = ) = Vr( ν S = ) = σ =, pz ( )( ν pz ( )), (.B) where he rnsiion probbiliies p z ) nd qz ( ) my hve eiher Probi or Logisic specificion. ( Wihin our frmework, he discree-vlued erm ν is correled wih ε in equion (). We ssume h while S my be endogenous, S j, j > is uncorreled wih ε. Agin, z is ssumed o be predeermined or exogenous. Under hese ssumpions, he exogenous componen of S cn be consruced s c + λ S while he endogenous componen cn be consruced s v = S c λs. 9

11 By defining ν ν =, where σ ν, S σ is he sndrd deviion of ν condiionl on ν, S S, we consider he following join disribuion of ν nd ε : ν ρ σ S ~(, Σ ), Σ = ε ρ σs σ S () A key o developing unbised esimion of he model is in rewriing he shocks using he Cholesky decomposiion of he vrince covrince mrix ( Σ ): AA' Σ = () where A = ρ σ S -ρ σ S. Then, if ω is discree rndom vrible wih men zero nd vrince one nd if ω is sndrd Norml rndom vrible independen of ω, we cn rewrie ν nd ε s follows: ν = σ ω, ν, S ρσ ε ρ σ ω ρ σ ω ν ρ σ ω S = S + S = + σν, S S (3) Using equion (3), equion () cn be rewrien s: y σ ω ' S = S + x S + ν + ρ σsω σν, S, S ν ρσ, = qz ( )( qz ( ))( S ) + pz ( )( pz ( )) S, i.i.d. N(, ), (4)

12 where ν = S c λs. In he bove rnsformed model in (4), noe h he new disurbnce erms ω is uncorreled wih ny of he explnory vribles in he men equion. This llows us o employ he usul Hmilon s (989) pproch for consisen esimion of he model. In equion (4), noe h he erm ν serves s conrol vrible for he endogenous componen of S, llowing he coefficiens nd o cpure he effec of he exogenous componen of S. Idenificion requires h we cn idenify S v = c+ λs s he exogenous componen of S. This is chieved in wo wys. Firs, if n exogenous or predeermined z exiss, s in he Mrkov-swiching models wih ime-vrying rnsiion probbiliies, he model is idenified. Second, even when = 3 =, s in he Mrkov-swiching model wih fixed rnsiion probbiliies, he model is sill idenified so long s S is serilly correled nd S is exogenous, nd S serves s n insrumen for S. This is rue even hough S is unobserved. 3.3 Tesing for n Endogenous Se Vrible Esimion of he models in (7) nd (4) provide srighforwrd es for endogeneiy in S. Given vlid insrumens, es of he null hypohesis of no endogeneiy cn be performed s es of he null hypohesis h ρ = in (7) or ρ = in (4). Noe h his cn be inerpreed s n pplicion of he Wu (973) es for endogenous regressors, where in his cse he regressor is n unobserved len se vrible. 4. Mone Crlo Assessmen of he Proposed Bis Correcion Procedure To evlue he proposed bis correcion procedures we perform series of Mone Crlo experimens. In ech Mone Crlo experimen, d ses re genered from he model

13 given in equions (), (4), nd (5). To model correlion beween S nd ε we ssume h η in equion (4) nd ε hve he join norml disribuion given in (5). Noe h his ssumpion is he sme s h used o develop he bis correcion procedure bsed on (7) described in Secion 3., mening his bis correcion procedure will be opiml. The Mone Crlo experimens re clibred s follows: Ech experimen is performed for hree vlues of he correlion prmeer ρ =.3,.6 nd.9 nd wo smple sizes, T = nd 5. The vecor z is ssumed o be sclr nd is genered s n independen sndrd norml rndom vrible. The vecor x is lso ssumed o be sclr nd is genered o be exogenous, bu correled wih S. We chieve his by genering x s sndrd norml rndom vrible dded o he exogenous pr of S, which from equion (4) is equl o + S + z ' + z ' S. This yields correlions beween x nd S rnging from.5 o 3.4, depending on he prmeerizion of equion (5) used in he Mone Crlo experimen. Equion (4) is clibred wih =, =, =, =, σ =. 5 nd σ =. We consider wo differen ses of prmeers for he se process (4). These re: DGP (Time-Vrying Trnsiion Probbiliies): =.5, =, =.5, 3 = DGP (Fixed Trnsiion Probbiliies): =.5, =, =, 3 = Bsed on his clibrion we perform hree ses of Mone Crlo experimens. In he firs, d is genered from DGP nd he model is esimed wihou bis correcion. This experimen will verify he bis in he regression prmeers discussed in secion. In he second nd hird experimens, d is genered from DGP nd DGP, nd boh of he bis correcion procedures discussed in secion 3, h bsed on equion (7) nd h bsed on equion (4), re implemened. In ech cse, he likelihood funcion for he pproprie model is consruced

14 using he filer given in Hmilon (989) nd mximized compuionlly. The resuls of he Mone Crlo experimens re conined in Tbles -4. Ech ble shows he men nd sndrd deviion of he mximum likelihood poin esimes of he prmeers of () nd (4), which will henceforh be referred o s he men nd sndrd deviion. Tble presens he resuls when d is genered ccording o DGP, nd no bis correcion procedure is employed. The ble demonsres h here is significn bis inroduced ino he regime-swiching coefficiens i nd i by he endogenous S. For exmple, when T = nd ρ =. 6, he men esimes of nd re over 3 sndrd deviions from he rue vlues. The bis is somewh smller in he esimes of nd, he men esimes of which re round one sndrd deviion from he rue vlues. As would be expeced, he bis in boh ses of prmeers is lrger when ρ =. 9 nd smller when ρ =.3. The bis is no miiged s he smple size grows when T = 5, he men esimes of i nd i re nerly he sme s for he T = cse. In Tble, we gin genere d ccording o DGP, bu now esime he model employing he bis correcion procedures from equion (7) (Tble ) nd equion (4) (Tble b). DGP supplies us wih informion o idenify he model prmeers boh from he ime vrying rnsiion probbiliies, z, nd he seril dependence of he se, S. The Mone Crlo resuls sugges h boh bis correcion procedures re very effecive. In priculr, he men esimes of i nd i re now close o he rue vlues for boh smple sizes considered. Tble 3 holds he resuls when he bis correcion echnique is pplied o DGP, in which here re no observed vribles wih which o idenify he model prmeers. Insed, only he unobserved lgged se, S, is vilble for idenificion. Tble 3 shows h for boh bis 3

15 correcion procedures, he esimes of i nd i re very close o he rue vlues. The sndrd deviions of he esimes in Tbles 3 re in generl somewh lrger hn hose in Tble. This is no surprising, s in Tble (bsed on DGP) here is more informion o idenify he prmeers, in he form of z, hn in Tbles 3. Overll, he Mone Crlo experimens confirm h ) endogenous regime swiching cn inroduce significn bis ino he mximum likelihood esimes of Mrkov regime-swiching model nd ) he bis correcion procedures proposed in Secion 3 re quie effecive in elimining his bis. In he nex secion we urn o n empiricl pplicion of he bis correcion procedures. 5. An Applicion: Mesuremen Error nd Esimion of he Voliliy Feedbck Effec A sylized fc of U.S. equiy reurn d is h he voliliy of relized reurns is imevrying nd predicble. Given his, clssic porfolio heory would imply h he equiy risk premium, he expeced reurn of he mrke porfolio bove he risk-free re, should lso be ime-vrying nd respond posiively o he expecion of fuure voliliy. However, he d sugges h relized reurns nd relized voliliy, s mesured by squred reurns, re negively correled. One explnion for he observed d is h while invesors do require n increse in expeced reurn in exchnge for expeced fuure voliliy, hey re lso ofen surprised by news bou relized voliliy. This voliliy feedbck effec crees reducion in prices in he period in which he increse in voliliy is relized. If he voliliy feedbck effec is srong enough, i my cree negive conemporneous correlion beween relized reurns nd voliliy in he d. The voliliy feedbck effec hs been invesiged exensively in he 4

16 lierure, see for exmple French, Schwer nd Smbugh (987), Turner, Srz nd Nelson (989), Cmpbell nd Henschell (99), Beker nd Wu () nd Kim, Morley nd Nelson (). One pproch o modeling ime-vrying voliliy is hrough he use of regimeswiching model. For exmple, Turner, Srz nd Nelson (989), henceforh TSN, model he excess of equiy reurns over he risk free re, r, s rising from norml disribuion wih ime dependen expecion µ nd vrince: σ r, = σ r, ( S ) + σ r, S, S = (,) σ r, > σ r, (5) S follows firs-order Mrkov-swiching process wih fixed rnsiion probbiliies, h is, he process in (4) wih =. The rnsiion probbiliies re hen given by: 3 = P( S P( S = S = ) = q = P( η ( ) = Φ( ) = S = ) = p = P( η > ( ( + ))) = Φ( ( + )) In his model, excess reurns re drwn from eiher high vrince disribuion ( S = ) or low vrince disribuion ( S = ). Also, since he se vrible S is persisen, deviions of he reurn vrince bove is uncondiionl men re predicble. We would hus expec risk premium h vries wih he predicble componen of he vrince. This cn be modeled s follows: µ = θ + θ PS ( Ψ ) (6) 5

17 Here, P ) mesures he probbiliy of he high vrince se ime given n ( S Ψ informion se ded ime -. TSN ssume h Ψ includes ll ps reurns, r...r nd P ( ) is hus equivlen o he filered probbiliy of he se. In (6), θ is equl o he S Ψ risk premium when here is no probbiliy plced on he occurrence of he high-vrince se. Assuming h θ is posiive, he risk premium is incresing in he probbiliy of he highvrince se. The model bove does no incorpore voliliy feedbck in deermining he cul excess reurn. TSN incorpore voliliy feedbck by modeling r s follows: r = µ + θ ( P( S Ψ ) P( S Ψ )) + ζ ~ N(, σς, ) σ ς, = σ ς, ( S ) + σ ς, S ζ (7) The model in (7) cn be moived s follows. A he beginning of period, he risk premium µ is deermined bsed on he expecion of he high-vrince se in period bsed on - informion. During period ddiionl informion regrding he incidence of he high-vrince se is observed. By he end of period, his informion cn be colleced in he informion se Ψ. When P S Ψ ) P( S ), informion reveled during he period hs surprised ( Ψ gens regrding he occurrence of he high-vrince se. If θ <, surprises h revel higher probbiliy of he high-vrince se re viewed negively by invesors, nd hus reduce he conemporneous reurn. One difficuly in esiming he bove model of voliliy feedbck is h here exiss discrepncy beween he invesors informion se nd he economericin s d se. In 6

18 priculr, while Ψ my be summrized by ll d up o -, h is Ψ = { r, r,...}, he informion se Ψ includes informion h is no summrized in he resercher s d se on observed reurns. This is becuse, while he resercher s d se is colleced discreely he beginning of ech period, he mrke pricipns coninuously observe rdes h occur during he period. This is priculr problem when he period is relively long, s is he cse for he monhly d se used by TSN. To hndle his esimion difficuly, TSN use he cul se, S, s proxy for P( Ψ ). Th is, hey esime: S r = µ + θ ( S P( S Ψ )) + ζ ~ N(, σ ) ς, σ ς, = σ ( S, ) + ς σ S ς, ζ (8) This pproch inroduces mesuremen error ino he explnory vribles of he esimed equion, s S is noisy mesure of PS Ψ. In priculr, noe h he residul in (8) is ( ) ζ = ζ + θ ( P( S Ψ ) S ) so h S will be correled wih he residul ζ in equion (8). The resuls from secion sugges h mximum likelihood esimes of he prmeers of equion (8) will hen be bised. The bis correcion echniques inroduced in Secion 3 cn be used o correc for his bis. We show his here using he echnique of Secion 3., in which join normliy ssumpion is employed. As here re no ime-vrying rnsiion probbiliies in he TSN model, h is z =, here is single insrumen in he model, S. Thus idenificion requires h S be Mrkov- 7

19 swiching, rher hn independen swiching. Formlly, equion (7) for he model in (8) becomes: r = µ + θ ( S P( S ψ )) + ω = ς ρσ ς,m ( S ) ρσ ρσ ς,m ( S ) + ς,m S ρσ ς,m S + ω (9) We esime he model in (9) using monhly reurns for vlue-weighed porfolio of ll NYSE-lised socks in excess of he one-monh Tresury Bill re, our mesure of he risk free re, over he smple period 95-. This is he sme d s used in Kim, Morley nd Nelson (). The firs pnel of Tble 4 shows hese esimes when no bis correcion is employed, h is ρ =. These esimes, which re similr o hose in TSN, re consisen wih boh posiive relionship beween he risk premium nd expeced fuure voliliy ( θ > nd θ > ) nd subsnil voliliy feedbck effec ( θ << ). The esimes re lso consisen wih dominn voliliy feedbck effec, h is θ is very smll relevn o θ. The second pnel shows he esimes when bis correcion is employed, so h ρ is esimed. There is subsnil evidence of endogeneiy, s ρ is esimed o be lrge in bsolue vlue nd sisiclly significn. Alhough he prmeer esimes re very imprecise, correcing for his endogeneiy does yield subsnil chnges in he poin esimes of he model prmeers. Firs, θ is esimed o be nerly en imes he esime for he model wih no bis correcion, suggesing risk premium h is much more responsive o forecsed fuure voliliy. Second, θ is esimed o be hlf he esime for he model wih no bis correcion, suggesing srong voliliy feedbck effec, bu less so hn for he model wih no bis correcion. 6. Conclusion 8

20 We hve explored he implicions of relxing he ssumpion of exogeneiy for he len se vrible in Mrkov regime-swiching regression. We show h endogeneiy of he se vrible cn led o significn esimion bis nd hve considered wo echniques o correc for his esimion bis. The firs exends he endogenous swiching models in Mddl nd Nelson (975) o he Hmilon (989) regime swiching regression, in which he swiching process is serilly dependen. This pproch relies on join normliy ssumpion for he relionship beween he regression error erms nd he innovions o he se relizion equion. The second echnique is bsed on he inerpreion of he endogenous swiching regression s regression model wih endogenous unobserved dummy vribles. In his cse he regression cn be esimed using insrumenl vribles echniques. For boh echniques, idenificion cn be chieved using informion from exernl, exogenous vribles h ffec he rnsiion probbiliies of he regime-swiching process. However, even in he fixed rnsiion probbiliy cse, we hve shown h idenificion is chieved if he se process is serilly dependen, nd lgged vlues of he se re exogenous. This is rue even hough he lgged se is unobserved. Mone Crlo experimens confirm h he proposed bis-correcion procedure is quie good for empiriclly relevn d genering processes. We pply he echnique o he voliliy feedbck model of equiy reurns given in Turner, Srz nd Nelson (989). 9

21 References Beker, G. nd G. Wu,, Asymmeric Voliliy nd Risk in Equiy Mrkes, Review of Finncil Sudies, 3, -4. Cmpbell, J. Y. nd L. Henschel, 99, No News is Good News: An Asymmeric Model of Chnging Voliliy in Sock Reurns, Journl of Finncil Economics, 3, Diebold, F., Lee, J-H. nd G. Weinbch, 994, Regime Swiching wih Time-Vrying Trnsiion Probbiliies, in Non-sionry Time Series Anlysis nd Coinegrion, ed. C. Hrgreves, Oxford Universiy Press, Oxford, U.K. Filrdo, A.J., 994, Business-Cycle Phses nd Their Trnsiionl Dynmics, Journl of Business nd Economic Sisics,, French, K.R., Schwer, G. nd R. F. Smbugh, 987, Expeced Sock Reurns nd Voliliy, Journl of Finncil Economics, 9, 3-9. Goldfeld, S.M. nd R.E. Qund, 97, in Nonliner Mehods in Economerics, Norh Hollnd, Amserdm. Goldfeld, S.M. nd R.E. Qund, 973, A Mrkov Model for Swiching Regressions, Journl of Economerics,, 3-6. Hmilon, J.D., 989, A New Approch o he Economic Anlysis of Nonsionry Time Series nd he Business Cycle, Economeric, 57, Hmilon, J.D., 99, Anlysis of Time Series Subjec o Chnges in Regime, Journl of Economerics, 45, Kim, C.J., Morley, J.C. nd C.R. Nelson,, Is There Posiive Relionship Beween Sock Mrke Voliliy nd he Equiy Premium?, Journl of Money, Credi nd Bnking, forhcoming. Mddl, G.S. nd F. Nelson, 975, Swiching Regression Models wih Exogenous nd Endogenous Swiching, Proceedings of he Americn Sisicl Associion, Mddl, G.S., 984, Disequilibrium, Self-Selecion nd Swiching Models, in Hndbook of Economerics, eds. Z. Griliches nd M.D. Inrilligor, Norh Hollnd, Amserdm. Owyng, M.,, Modeling Volcker s Non-Absorbing Se: Agnosic Idenificion of Mrkov-Swiching VAR, working pper, Federl Reserve Bnk of S. Louis. Qund, R.E., 958, The Esimion of he Prmeers of Liner Regression Sysem Obeying Two Sepre Regimes, Journl of he Americn Sisicl Associion, 53,

22 Qund, R.E., 97, A New Approch o Esiming Swichign Regressions, Journl of he Americn Sisicl Associion, 67, Sims, C. nd T. Zh,, Mcroeconomic Swiching, mimeo. Turner, C. M., Srz, R. nd C.R. Nelson, 989, A Mrkov Model of Heeroskedsiciy, Risk, nd Lerning in he Sock Mrke, Journl of Finncil Economics, 5, 3-. Wu, D., 973, Alernive Tess of Independence Beween Sochsic Regressors nd Disurbnces, Economeric, 4,

23 Tble Mone Crlo : Mximum Likelihood Esimion of Time-Vrying Trnsiion Probbiliy Mrkov-Swiching Regression Model (TVP-MS) No Bis Correcion Prmeer T= ρ =.3 ρ =. 6 ρ =. 9 Prmeer Esimes: Men / Sd. Dev. Prmeer Esimes: Men / Sd. Dev. Prmeer Esimes: Men / Sd. Dev. =.9 /.5.83 /.5.74 /.5 = -.77 / / /. =.97 /.4.95 /.4.93 /.4 = -.6 /.7 -. / /.6 = / / /.6 =.3 /.8.6 /.8.7 /.9 =.5.54 /.9.56 /.8.56 /.9 3 =.3 /.5.3 /.53. /.5 T=5 =.9 /.4.8 /.3.74 /.3 = -.78 / / /.6 =.98 /..95 /..93 /. = -.6 /.4 -. / /.4 = /. -.5 /. -.5 /. =. /.7.3 /.7.4 /.7 =.5.5 /..53 /..54 /. 3 =.3 /.5.4 /.7.5 /.9 Noes: Ech column conins he men nd sndrd deviion of he mximum likelihood poin esimes from he Mone Crlo experimen.

24 Tble Mone Crlo : Mximum Likelihood Esimion of Time-Vrying Trnsiion Probbiliy Mrkov-Swiching Regression Model (TVP-MS) Bis Correcion from Equion (7) Prmeer T= ρ =.3 ρ =. 6 ρ =. 9 Prmeer Esimes: Men / Sd. Dev. Prmeer Esimes: Men / Sd. Dev. Prmeer Esimes: Men / Sd. Dev. =. /.9. /.9. /.7 = -.98 / / /.6 =. /.4.99 /.4.99 /.4 = -. /.9 -. /.8 -. /.7 = / / /.4 =. /.9. /.7.3 /. =.5.53 /.8.5 /.7.5 /.4 3 =. /.48. /.44.9 /.37 T=5 =.99 /.6.99 /.5. /.4 = -.98 / / /. =. /.3. /.. /. = -. /.5 -. /.5 -. /.4 = / / /.8 =. /.7. /.6. /.3 =.5.5 /..5 /..5 /.8 3 =.4 /.5.3 /.4. /.9 Noes: Ech column conins he men nd sndrd deviion of he mximum likelihood poin esimes from he Mone Crlo experimen. 3

25 Tble b Mone Crlo : Mximum Likelihood Esimion of Time-Vrying Trnsiion Probbiliy Mrkov-Swiching Regression Model (TVP-MS) Bis Correcion from Equion (4) Prmeer T= ρ =.3 ρ =. 6 ρ =. 9 Prmeer Esimes: Men / Sd. Dev. Prmeer Esimes: Men / Sd. Dev. Prmeer Esimes: Men / Sd. Dev. =.99 /..99 /.. /. = -.98 / / /.3 =. /.4.99 /.4.99 /.4 = -. /.9 -. / /.8 = / / /.5 =. /.8. /.7.97 /.4 =.5.5 /.8.53 /.7.5 /.5 3 =. /.47.5 /.44.9 /.4 T=5 =.98 /.6.98 /.6.99 /.6 = -.96 / / /.5 =. /.3.99 /.3.99 /. = -. / / /.5 = / / /.9 =. /.7.98 /.5.94 /.4 =.5.5 /..5 /..48 /.9 3 =.4 /.7.97 /.3.9 /. Noes: Ech column conins he men nd sndrd deviion of he mximum likelihood poin esimes from he Mone Crlo experimen. 4

26 Tble 3 Mone Crlo 3: Mximum Likelihood Esimion of Fixed Trnsiion Probbiliy Mrkov- Swiching Regression Model (FTP-MS) Bis Correcion from Equion (7) Prmeer T= ρ =.3 ρ =. 6 ρ =. 9 Prmeer Esimes: Men / Sd. Dev. Prmeer Esimes: Men / Sd. Dev. Prmeer Esimes: Men / Sd. Dev. =. /.4. /.3. /.9 = -. /.8 -. / /.7 =.99 /.5. /.5.99 /.4 = -. /. -. /.9 -. /.7 = / / /.4 =.98 /..99 /.. /. = = T=5 =. /.8. /.8. /.6 = -.99 / / /. =. /.3.99 /.3.99 /. = -. /.6 -. /.6 -. /.5 = / / /.9 =. /.4.99 /.4.99 /.3 = = Noes: Ech column conins he men nd sndrd deviion of he mximum likelihood poin esimes from he Mone Crlo experimen. 5

27 Tble 3b Mone Crlo 3: Mximum Likelihood Esimion of Fixed Trnsiion Probbiliy Mrkov- Swiching Regression Model (FTP-MS) Bis Correcion from Equion (4) Prmeer T= ρ =.3 ρ =. 6 ρ =. 9 Prmeer Esimes: Men / Sd. Dev. Prmeer Esimes: Men / Sd. Dev. Prmeer Esimes: Men / Sd. Dev. =. /.5. /.5.6 /.6 = -. / /.3 -. /.3 =.99 /.5.99 /.5.99 /.4 = -. /. -. /. -. /.8 = / / /.5 =.99 /..98 /..95 /. = = T=5 =. /.9. /.9.4 /.9 = -. /.7 -. / /.7 =. /.3.99 /.3.99 /. = -. /.6 -. /.6 -. /.5 = / / /. =.98 /.4.98 /.4.96 /.4 = = Noes: Ech column conins he men nd sndrd deviion of he mximum likelihood poin esimes from he Mone Crlo experimen. 6

28 Tble 4 Mximum Likelihood Esimes of Voliliy-Feedbck Model from Turner, Srz nd Nelson (989) (sndrd errors in prenhesis) Prmeer Wihou Bis Correcion Wih Bis Correcion θ.6 (.6) θ. (.3) θ -. (.5) σ ς,.38 (.) σ ς,.67 (.4) -. (.5) 3.5 (.7).4 (.7).9 (.34) -. (.49).4 (.).74 (.6) -.6 (.6) 3.6 (.3) ρ (.9) 7

Estimation of Markov Regime-Switching Regression Models with Endogenous Switching

Estimation of Markov Regime-Switching Regression Models with Endogenous Switching WORKING PAPER ERIE Esimion of Mrkov Regime-wiching Regression Models wih Endogenous wiching Chng-Jin Kim Jerem Piger nd Richrd r Working Pper 3-5B hp://reserch.slouisfed.org/wp/3/3-5.pdf June 3 Revised

More information

A new model for limit order book dynamics

A new model for limit order book dynamics Anewmodelforlimiorderbookdynmics JeffreyR.Russell UniversiyofChicgo,GrdueSchoolofBusiness TejinKim UniversiyofChicgo,DeprmenofSisics Absrc:Thispperproposesnewmodelforlimiorderbookdynmics.Thelimiorderbookconsiss

More information

A Kalman filtering simulation

A Kalman filtering simulation A Klmn filering simulion The performnce of Klmn filering hs been esed on he bsis of wo differen dynmicl models, ssuming eiher moion wih consn elociy or wih consn ccelerion. The former is epeced o beer

More information

e t dt e t dt = lim e t dt T (1 e T ) = 1

e t dt e t dt = lim e t dt T (1 e T ) = 1 Improper Inegrls There re wo ypes of improper inegrls - hose wih infinie limis of inegrion, nd hose wih inegrnds h pproch some poin wihin he limis of inegrion. Firs we will consider inegrls wih infinie

More information

3. Renewal Limit Theorems

3. Renewal Limit Theorems Virul Lborories > 14. Renewl Processes > 1 2 3 3. Renewl Limi Theorems In he inroducion o renewl processes, we noed h he rrivl ime process nd he couning process re inverses, in sens The rrivl ime process

More information

September 20 Homework Solutions

September 20 Homework Solutions College of Engineering nd Compuer Science Mechnicl Engineering Deprmen Mechnicl Engineering A Seminr in Engineering Anlysis Fll 7 Number 66 Insrucor: Lrry Creo Sepember Homework Soluions Find he specrum

More information

Trading Collar, Intraday Periodicity, and Stock Market Volatility. Satheesh V. Aradhyula University of Arizona. A. Tolga Ergun University of Arizona

Trading Collar, Intraday Periodicity, and Stock Market Volatility. Satheesh V. Aradhyula University of Arizona. A. Tolga Ergun University of Arizona Trding Collr, Inrdy Periodiciy, nd Sock Mrke Voliliy Sheesh V. Ardhyul Universiy of Arizon A. Tolg Ergun Universiy of Arizon My, 00 Absrc: Using 5 minue d, we exmine mrke voliliy in he Dow Jones Indusril

More information

Contraction Mapping Principle Approach to Differential Equations

Contraction Mapping Principle Approach to Differential Equations epl Journl of Science echnology 0 (009) 49-53 Conrcion pping Principle pproch o Differenil Equions Bishnu P. Dhungn Deprmen of hemics, hendr Rn Cmpus ribhuvn Universiy, Khmu epl bsrc Using n eension of

More information

4.8 Improper Integrals

4.8 Improper Integrals 4.8 Improper Inegrls Well you ve mde i hrough ll he inegrion echniques. Congrs! Unforunely for us, we sill need o cover one more inegrl. They re clled Improper Inegrls. A his poin, we ve only del wih inegrls

More information

Probability, Estimators, and Stationarity

Probability, Estimators, and Stationarity Chper Probbiliy, Esimors, nd Sionriy Consider signl genered by dynmicl process, R, R. Considering s funcion of ime, we re opering in he ime domin. A fundmenl wy o chrcerize he dynmics using he ime domin

More information

Minimum Squared Error

Minimum Squared Error Minimum Squred Error LDF: Minimum Squred-Error Procedures Ide: conver o esier nd eer undersood prolem Percepron y i > 0 for ll smples y i solve sysem of liner inequliies MSE procedure y i i for ll smples

More information

AJAE appendix for Is Exchange Rate Pass-Through in Pork Meat Export Prices Constrained by the Supply of Live Hogs?

AJAE appendix for Is Exchange Rate Pass-Through in Pork Meat Export Prices Constrained by the Supply of Live Hogs? AJAE ppendix for Is Exchnge Re Pss-Through in Por Me Expor Prices Consrined by he Supply of Live Hogs? Jen-Philippe Gervis Cnd Reserch Chir in Agri-indusries nd Inernionl Trde Cener for Reserch in he Economics

More information

Chapter 2: Evaluative Feedback

Chapter 2: Evaluative Feedback Chper 2: Evluive Feedbck Evluing cions vs. insrucing by giving correc cions Pure evluive feedbck depends olly on he cion ken. Pure insrucive feedbck depends no ll on he cion ken. Supervised lerning is

More information

Minimum Squared Error

Minimum Squared Error Minimum Squred Error LDF: Minimum Squred-Error Procedures Ide: conver o esier nd eer undersood prolem Percepron y i > for ll smples y i solve sysem of liner inequliies MSE procedure y i = i for ll smples

More information

f t f a f x dx By Lin McMullin f x dx= f b f a. 2

f t f a f x dx By Lin McMullin f x dx= f b f a. 2 Accumulion: Thoughs On () By Lin McMullin f f f d = + The gols of he AP* Clculus progrm include he semen, Sudens should undersnd he definie inegrl s he ne ccumulion of chnge. 1 The Topicl Ouline includes

More information

The solution is often represented as a vector: 2xI + 4X2 + 2X3 + 4X4 + 2X5 = 4 2xI + 4X2 + 3X3 + 3X4 + 3X5 = 4. 3xI + 6X2 + 6X3 + 3X4 + 6X5 = 6.

The solution is often represented as a vector: 2xI + 4X2 + 2X3 + 4X4 + 2X5 = 4 2xI + 4X2 + 3X3 + 3X4 + 3X5 = 4. 3xI + 6X2 + 6X3 + 3X4 + 6X5 = 6. [~ o o :- o o ill] i 1. Mrices, Vecors, nd Guss-Jordn Eliminion 1 x y = = - z= The soluion is ofen represened s vecor: n his exmple, he process of eliminion works very smoohly. We cn elimine ll enries

More information

The Taiwan stock market does follow a random walk. Abstract

The Taiwan stock market does follow a random walk. Abstract The Tiwn soc mre does follow rndom wl D Bue Loc Feng Chi Universiy Absrc Applying he Lo nd McKinly vrince rio es on he weely reurns from he Tiwn soc mre from 990 o mid 006, I obined resuls srongly indicive

More information

Motion. Part 2: Constant Acceleration. Acceleration. October Lab Physics. Ms. Levine 1. Acceleration. Acceleration. Units for Acceleration.

Motion. Part 2: Constant Acceleration. Acceleration. October Lab Physics. Ms. Levine 1. Acceleration. Acceleration. Units for Acceleration. Moion Accelerion Pr : Consn Accelerion Accelerion Accelerion Accelerion is he re of chnge of velociy. = v - vo = Δv Δ ccelerion = = v - vo chnge of velociy elpsed ime Accelerion is vecor, lhough in one-dimensionl

More information

1 jordan.mcd Eigenvalue-eigenvector approach to solving first order ODEs. -- Jordan normal (canonical) form. Instructor: Nam Sun Wang

1 jordan.mcd Eigenvalue-eigenvector approach to solving first order ODEs. -- Jordan normal (canonical) form. Instructor: Nam Sun Wang jordnmcd Eigenvlue-eigenvecor pproch o solving firs order ODEs -- ordn norml (cnonicl) form Insrucor: Nm Sun Wng Consider he following se of coupled firs order ODEs d d x x 5 x x d d x d d x x x 5 x x

More information

A Time Truncated Improved Group Sampling Plans for Rayleigh and Log - Logistic Distributions

A Time Truncated Improved Group Sampling Plans for Rayleigh and Log - Logistic Distributions ISSNOnline : 39-8753 ISSN Prin : 347-67 An ISO 397: 7 Cerified Orgnizion Vol. 5, Issue 5, My 6 A Time Trunced Improved Group Smpling Plns for Ryleigh nd og - ogisic Disribuions P.Kvipriy, A.R. Sudmni Rmswmy

More information

Convergence of Singular Integral Operators in Weighted Lebesgue Spaces

Convergence of Singular Integral Operators in Weighted Lebesgue Spaces EUROPEAN JOURNAL OF PURE AND APPLIED MATHEMATICS Vol. 10, No. 2, 2017, 335-347 ISSN 1307-5543 www.ejpm.com Published by New York Business Globl Convergence of Singulr Inegrl Operors in Weighed Lebesgue

More information

0 for t < 0 1 for t > 0

0 for t < 0 1 for t > 0 8.0 Sep nd del funcions Auhor: Jeremy Orloff The uni Sep Funcion We define he uni sep funcion by u() = 0 for < 0 for > 0 I is clled he uni sep funcion becuse i kes uni sep = 0. I is someimes clled he Heviside

More information

USING ITERATIVE LINEAR REGRESSION MODEL TO TIME SERIES MODELS

USING ITERATIVE LINEAR REGRESSION MODEL TO TIME SERIES MODELS Elecronic Journl of Applied Sisicl Anlysis EJASA (202), Elecron. J. App. S. Anl., Vol. 5, Issue 2, 37 50 e-issn 2070-5948, DOI 0.285/i20705948v5n2p37 202 Universià del Sleno hp://sib-ese.unile.i/index.php/ejs/index

More information

THREE IMPORTANT CONCEPTS IN TIME SERIES ANALYSIS: STATIONARITY, CROSSING RATES, AND THE WOLD REPRESENTATION THEOREM

THREE IMPORTANT CONCEPTS IN TIME SERIES ANALYSIS: STATIONARITY, CROSSING RATES, AND THE WOLD REPRESENTATION THEOREM THR IMPORTANT CONCPTS IN TIM SRIS ANALYSIS: STATIONARITY, CROSSING RATS, AND TH WOLD RPRSNTATION THORM Prof. Thoms B. Fomb Deprmen of conomics Souhern Mehodis Universi June 8 I. Definiion of Covrince Sionri

More information

MATH 124 AND 125 FINAL EXAM REVIEW PACKET (Revised spring 2008)

MATH 124 AND 125 FINAL EXAM REVIEW PACKET (Revised spring 2008) MATH 14 AND 15 FINAL EXAM REVIEW PACKET (Revised spring 8) The following quesions cn be used s review for Mh 14/ 15 These quesions re no cul smples of quesions h will pper on he finl em, bu hey will provide

More information

1.0 Electrical Systems

1.0 Electrical Systems . Elecricl Sysems The ypes of dynmicl sysems we will e sudying cn e modeled in erms of lgeric equions, differenil equions, or inegrl equions. We will egin y looking fmilir mhemicl models of idel resisors,

More information

An integral having either an infinite limit of integration or an unbounded integrand is called improper. Here are two examples.

An integral having either an infinite limit of integration or an unbounded integrand is called improper. Here are two examples. Improper Inegrls To his poin we hve only considered inegrls f(x) wih he is of inegrion nd b finie nd he inegrnd f(x) bounded (nd in fc coninuous excep possibly for finiely mny jump disconinuiies) An inegrl

More information

Dynamic Relatedness Analysis of Three Exchange Rate Markets Volatility: Study of Korea, Taiwan and Thailand

Dynamic Relatedness Analysis of Three Exchange Rate Markets Volatility: Study of Korea, Taiwan and Thailand Inernionl Conference on Advnced Compuer Science nd Elecronics Informion (ICACSEI 3) Dynmic Reledness Anlysis of Tree Excnge Re Mrkes Voliliy: Sudy of Kore, Tiwn nd Tilnd Wnn-Jyi Horng, Tien-Cung Hu, Deprmen

More information

Interest Rate Determination : An Error Correction Model. Ila Patnaik Deepa Vasudevan

Interest Rate Determination : An Error Correction Model. Ila Patnaik Deepa Vasudevan Ineres Re Deerminion : An Error Correcion Model Il Pnik Deep Vsudevn Nionl Council of Applied Economic Reserch 11, IP Ese New Delhi 110 002 Absrc Inegrion of he domesic mrke for funds wih foreign money

More information

Asymmetry in the Business Model: Revisiting the Friedman Plucking Model

Asymmetry in the Business Model: Revisiting the Friedman Plucking Model Insiue for Inernionl Economic Policy Working Pper eries Ellio chool of Inernionl Affirs The George Wshingon Universiy Asymmery in he Business odel: Revisiing he Friedmn Plucking odel IIEP WP 8 3 Tr inclir

More information

Average & instantaneous velocity and acceleration Motion with constant acceleration

Average & instantaneous velocity and acceleration Motion with constant acceleration Physics 7: Lecure Reminders Discussion nd Lb secions sr meeing ne week Fill ou Pink dd/drop form if you need o swich o differen secion h is FULL. Do i TODAY. Homework Ch. : 5, 7,, 3,, nd 6 Ch.: 6,, 3 Submission

More information

GENERALIZATION OF SOME INEQUALITIES VIA RIEMANN-LIOUVILLE FRACTIONAL CALCULUS

GENERALIZATION OF SOME INEQUALITIES VIA RIEMANN-LIOUVILLE FRACTIONAL CALCULUS - TAMKANG JOURNAL OF MATHEMATICS Volume 5, Number, 7-5, June doi:5556/jkjm555 Avilble online hp://journlsmhkueduw/ - - - GENERALIZATION OF SOME INEQUALITIES VIA RIEMANN-LIOUVILLE FRACTIONAL CALCULUS MARCELA

More information

REAL ANALYSIS I HOMEWORK 3. Chapter 1

REAL ANALYSIS I HOMEWORK 3. Chapter 1 REAL ANALYSIS I HOMEWORK 3 CİHAN BAHRAN The quesions re from Sein nd Shkrchi s e. Chper 1 18. Prove he following sserion: Every mesurble funcion is he limi.e. of sequence of coninuous funcions. We firs

More information

Volatility, Spillover Effects and Correlations in US and Major European Markets

Volatility, Spillover Effects and Correlations in US and Major European Markets Voliliy, Spillover Effecs nd Correlions in US nd Mjor Europen Mrkes Chrisos S. Svv Denise R. Osborn Len Gill Universiy of Mncheser Absrc This pper invesiges he rnsmission mechnism of price nd voliliy spillovers

More information

Nonlinear System Modelling: How to Estimate the. Highest Significant Order

Nonlinear System Modelling: How to Estimate the. Highest Significant Order IEEE Insrumenion nd Mesuremen Technology Conference nchorge,, US, - My Nonliner Sysem Modelling: ow o Esime he ighes Significn Order Neophyos Chirs, Ceri Evns nd Dvid Rees, Michel Solomou School of Elecronics,

More information

Asymmetry in the Business Cycle: Friedman s Plucking Model with Correlated Innovations

Asymmetry in the Business Cycle: Friedman s Plucking Model with Correlated Innovations Asymmery in he Business Cycle: Friedmn s Plucking odel ih Correled Innovions Tr. inclir Deprmen of Economics And he Ellio chool of Inernionl Affirs The George Wshingon Universiy Wshingon, DC 5 sinc@gu.edu

More information

22.615, MHD Theory of Fusion Systems Prof. Freidberg Lecture 9: The High Beta Tokamak

22.615, MHD Theory of Fusion Systems Prof. Freidberg Lecture 9: The High Beta Tokamak .65, MHD Theory of Fusion Sysems Prof. Freidberg Lecure 9: The High e Tokmk Summry of he Properies of n Ohmic Tokmk. Advnges:. good euilibrium (smll shif) b. good sbiliy ( ) c. good confinemen ( τ nr )

More information

Bias in Conditional and Unconditional Fixed Effects Logit Estimation: a Correction * Tom Coupé

Bias in Conditional and Unconditional Fixed Effects Logit Estimation: a Correction * Tom Coupé Bias in Condiional and Uncondiional Fixed Effecs Logi Esimaion: a Correcion * Tom Coupé Economics Educaion and Research Consorium, Naional Universiy of Kyiv Mohyla Academy Address: Vul Voloska 10, 04070

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

Journal of Mathematical Analysis and Applications. Two normality criteria and the converse of the Bloch principle

Journal of Mathematical Analysis and Applications. Two normality criteria and the converse of the Bloch principle J. Mh. Anl. Appl. 353 009) 43 48 Conens liss vilble ScienceDirec Journl of Mhemicl Anlysis nd Applicions www.elsevier.com/loce/jm Two normliy crieri nd he converse of he Bloch principle K.S. Chrk, J. Rieppo

More information

Transforms II - Wavelets Preliminary version please report errors, typos, and suggestions for improvements

Transforms II - Wavelets Preliminary version please report errors, typos, and suggestions for improvements EECS 3 Digil Signl Processing Universiy of Cliforni, Berkeley: Fll 007 Gspr November 4, 007 Trnsforms II - Wveles Preliminry version plese repor errors, ypos, nd suggesions for improvemens We follow n

More information

MAT 266 Calculus for Engineers II Notes on Chapter 6 Professor: John Quigg Semester: spring 2017

MAT 266 Calculus for Engineers II Notes on Chapter 6 Professor: John Quigg Semester: spring 2017 MAT 66 Clculus for Engineers II Noes on Chper 6 Professor: John Quigg Semeser: spring 7 Secion 6.: Inegrion by prs The Produc Rule is d d f()g() = f()g () + f ()g() Tking indefinie inegrls gives [f()g

More information

Observability of flow dependent structure functions and their use in data assimilation

Observability of flow dependent structure functions and their use in data assimilation Oserviliy of flow dependen srucure funcions nd heir use in d ssimilion Pierre Guhier nd Crisin Lupu Collorion wih Séphne Lroche, Mrk Buehner nd Ahmed Mhidji (Env. Cnd) 3rd meeing of he HORPEX DAOS-WG Monrél

More information

Solutions to Problems from Chapter 2

Solutions to Problems from Chapter 2 Soluions o Problems rom Chper Problem. The signls u() :5sgn(), u () :5sgn(), nd u h () :5sgn() re ploed respecively in Figures.,b,c. Noe h u h () :5sgn() :5; 8 including, bu u () :5sgn() is undeined..5

More information

1. Introduction. 1 b b

1. Introduction. 1 b b Journl of Mhemicl Inequliies Volume, Number 3 (007), 45 436 SOME IMPROVEMENTS OF GRÜSS TYPE INEQUALITY N. ELEZOVIĆ, LJ. MARANGUNIĆ AND J. PEČARIĆ (communiced b A. Čižmešij) Absrc. In his pper some inequliies

More information

5.1-The Initial-Value Problems For Ordinary Differential Equations

5.1-The Initial-Value Problems For Ordinary Differential Equations 5.-The Iniil-Vlue Problems For Ordinry Differenil Equions Consider solving iniil-vlue problems for ordinry differenil equions: (*) y f, y, b, y. If we know he generl soluion y of he ordinry differenil

More information

Chapter Direct Method of Interpolation

Chapter Direct Method of Interpolation Chper 5. Direc Mehod of Inerpolion Afer reding his chper, you should be ble o:. pply he direc mehod of inerpolion,. sole problems using he direc mehod of inerpolion, nd. use he direc mehod inerpolns o

More information

Bayesian Inference for Static Traffic Network Flows with Mobile Sensor Data

Bayesian Inference for Static Traffic Network Flows with Mobile Sensor Data Proceedings of he 51 s Hwii Inernionl Conference on Sysem Sciences 218 Byesin Inference for Sic Trffic Nework Flows wih Mobile Sensor D Zhen Tn Cornell Universiy z78@cornell.edu H.Oliver Go Cornell Universiy

More information

On Source and Channel Codes for Multiple Inputs and Outputs: Does Multiple Description Meet Space Time? 1

On Source and Channel Codes for Multiple Inputs and Outputs: Does Multiple Description Meet Space Time? 1 On Source nd Chnnel Codes for Muliple Inpus nd Oupus: oes Muliple escripion Mee Spce Time? Michelle Effros Rlf Koeer 3 Andre J. Goldsmih 4 Muriel Médrd 5 ep. of Elecricl Eng., 36-93, Cliforni Insiue of

More information

3D Transformations. Computer Graphics COMP 770 (236) Spring Instructor: Brandon Lloyd 1/26/07 1

3D Transformations. Computer Graphics COMP 770 (236) Spring Instructor: Brandon Lloyd 1/26/07 1 D Trnsformions Compuer Grphics COMP 770 (6) Spring 007 Insrucor: Brndon Lloyd /6/07 Geomery Geomeric eniies, such s poins in spce, exis wihou numers. Coordines re nming scheme. The sme poin cn e descried

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

Green s Functions and Comparison Theorems for Differential Equations on Measure Chains

Green s Functions and Comparison Theorems for Differential Equations on Measure Chains Green s Funcions nd Comprison Theorems for Differenil Equions on Mesure Chins Lynn Erbe nd Alln Peerson Deprmen of Mhemics nd Sisics, Universiy of Nebrsk-Lincoln Lincoln,NE 68588-0323 lerbe@@mh.unl.edu

More information

TIMELINESS, ACCURACY, AND RELEVANCE IN DYNAMIC INCENTIVE CONTRACTS

TIMELINESS, ACCURACY, AND RELEVANCE IN DYNAMIC INCENTIVE CONTRACTS TIMELINESS, ACCURACY, AND RELEVANCE IN DYNAMIC INCENTIVE CONTRACTS by Peer O. Chrisensen Universiy of Souhern Denmrk Odense, Denmrk Gerld A. Felhm Universiy of Briish Columbi Vncouver, Cnd Chrisin Hofmnn

More information

EXISTENCE AND UNIQUENESS OF SOLUTIONS FOR A SECOND-ORDER ITERATIVE BOUNDARY-VALUE PROBLEM

EXISTENCE AND UNIQUENESS OF SOLUTIONS FOR A SECOND-ORDER ITERATIVE BOUNDARY-VALUE PROBLEM Elecronic Journl of Differenil Equions, Vol. 208 (208), No. 50, pp. 6. ISSN: 072-669. URL: hp://ejde.mh.xse.edu or hp://ejde.mh.un.edu EXISTENCE AND UNIQUENESS OF SOLUTIONS FOR A SECOND-ORDER ITERATIVE

More information

ON NEW INEQUALITIES OF SIMPSON S TYPE FOR FUNCTIONS WHOSE SECOND DERIVATIVES ABSOLUTE VALUES ARE CONVEX

ON NEW INEQUALITIES OF SIMPSON S TYPE FOR FUNCTIONS WHOSE SECOND DERIVATIVES ABSOLUTE VALUES ARE CONVEX Journl of Applied Mhemics, Sisics nd Informics JAMSI), 9 ), No. ON NEW INEQUALITIES OF SIMPSON S TYPE FOR FUNCTIONS WHOSE SECOND DERIVATIVES ABSOLUTE VALUES ARE CONVEX MEHMET ZEKI SARIKAYA, ERHAN. SET

More information

A Simple Method to Solve Quartic Equations. Key words: Polynomials, Quartics, Equations of the Fourth Degree INTRODUCTION

A Simple Method to Solve Quartic Equations. Key words: Polynomials, Quartics, Equations of the Fourth Degree INTRODUCTION Ausrlin Journl of Bsic nd Applied Sciences, 6(6): -6, 0 ISSN 99-878 A Simple Mehod o Solve Quric Equions Amir Fhi, Poo Mobdersn, Rhim Fhi Deprmen of Elecricl Engineering, Urmi brnch, Islmic Ad Universi,

More information

Procedia Computer Science

Procedia Computer Science Procedi Compuer Science 00 (0) 000 000 Procedi Compuer Science www.elsevier.com/loce/procedi The Third Informion Sysems Inernionl Conference The Exisence of Polynomil Soluion of he Nonliner Dynmicl Sysems

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

Magnetostatics Bar Magnet. Magnetostatics Oersted s Experiment

Magnetostatics Bar Magnet. Magnetostatics Oersted s Experiment Mgneosics Br Mgne As fr bck s 4500 yers go, he Chinese discovered h cerin ypes of iron ore could rc ech oher nd cerin mels. Iron filings "mp" of br mgne s field Crefully suspended slivers of his mel were

More information

Physics 2A HW #3 Solutions

Physics 2A HW #3 Solutions Chper 3 Focus on Conceps: 3, 4, 6, 9 Problems: 9, 9, 3, 41, 66, 7, 75, 77 Phsics A HW #3 Soluions Focus On Conceps 3-3 (c) The ccelerion due o grvi is he sme for boh blls, despie he fc h he hve differen

More information

Process Monitoring and Feedforward Control for Proactive Quality Improvement

Process Monitoring and Feedforward Control for Proactive Quality Improvement Inernionl Journl of Performbiliy Engineering Vol. 8, No. 6, November 0, pp. 60-64. RAMS Consulns Prined in Indi Process Monioring nd Feedforwrd Conrol for Procive Quliy Improvemen. Inroducion LIHUI SHI

More information

S Radio transmission and network access Exercise 1-2

S Radio transmission and network access Exercise 1-2 S-7.330 Rdio rnsmission nd nework ccess Exercise 1 - P1 In four-symbol digil sysem wih eqully probble symbols he pulses in he figure re used in rnsmission over AWGN-chnnel. s () s () s () s () 1 3 4 )

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

FINANCIAL ECONOMETRICS

FINANCIAL ECONOMETRICS FINANCIAL ECONOMETRICS SPRING 7 WEEK VII MULTIVARIATE MODELLING OF VOLATILITY Prof. Dr. Burç ÜLENGİN MULTIVARIATE VOLATILITY Tere my e inercions mong e condiionl vrince of e reurn series. Also covrince

More information

Properties of Logarithms. Solving Exponential and Logarithmic Equations. Properties of Logarithms. Properties of Logarithms. ( x)

Properties of Logarithms. Solving Exponential and Logarithmic Equations. Properties of Logarithms. Properties of Logarithms. ( x) Properies of Logrihms Solving Eponenil nd Logrihmic Equions Properies of Logrihms Produc Rule ( ) log mn = log m + log n ( ) log = log + log Properies of Logrihms Quoien Rule log m = logm logn n log7 =

More information

PHYSICS 1210 Exam 1 University of Wyoming 14 February points

PHYSICS 1210 Exam 1 University of Wyoming 14 February points PHYSICS 1210 Em 1 Uniersiy of Wyoming 14 Februry 2013 150 poins This es is open-noe nd closed-book. Clculors re permied bu compuers re no. No collborion, consulion, or communicion wih oher people (oher

More information

An Analysis of the Relation Between the Stewardship and Valuation Roles of Earnings

An Analysis of the Relation Between the Stewardship and Valuation Roles of Earnings An Anlysis of he Relion Beween he Sewrdship nd Vluion Roles of rnings Rober Bushmn Kenn-Flgler Business School Universiy of Norh Crolin - Chpel Hill llen ngel * Abbie Smih Grdue School of Business The

More information

Forecasting Correlation and Covariance with a. Range-Based Dynamic Conditional Correlation Model

Forecasting Correlation and Covariance with a. Range-Based Dynamic Conditional Correlation Model Forecsing Correlion nd Covrince wih Rnge-Bsed Dynmic Condiionl Correlion Model Ry Y Chou * Insiue of Economics, Acdemi Sinic Nhn Liu Deprmen of Mngemen Science, Nionl Chio-ung Universiy Chun-Chou Wu Deprmen

More information

The average rate of change for continuous time models

The average rate of change for continuous time models Behvior Reserch Mehods 9, (), 68-78 doi:.78/brm...68 he verge re of chnge for coninuous ime models EN ELLE Universiy of Nore Dme, Nore Dme, Indin he verge re of chnge (ARC) is concep h hs been misundersood

More information

LAPLACE TRANSFORM OVERCOMING PRINCIPLE DRAWBACKS IN APPLICATION OF THE VARIATIONAL ITERATION METHOD TO FRACTIONAL HEAT EQUATIONS

LAPLACE TRANSFORM OVERCOMING PRINCIPLE DRAWBACKS IN APPLICATION OF THE VARIATIONAL ITERATION METHOD TO FRACTIONAL HEAT EQUATIONS Wu, G.-.: Lplce Trnsform Overcoming Principle Drwbcks in Applicion... THERMAL SIENE: Yer 22, Vol. 6, No. 4, pp. 257-26 257 Open forum LAPLAE TRANSFORM OVEROMING PRINIPLE DRAWBAKS IN APPLIATION OF THE VARIATIONAL

More information

A LIMIT-POINT CRITERION FOR A SECOND-ORDER LINEAR DIFFERENTIAL OPERATOR IAN KNOWLES

A LIMIT-POINT CRITERION FOR A SECOND-ORDER LINEAR DIFFERENTIAL OPERATOR IAN KNOWLES A LIMIT-POINT CRITERION FOR A SECOND-ORDER LINEAR DIFFERENTIAL OPERATOR j IAN KNOWLES 1. Inroducion Consider he forml differenil operor T defined by el, (1) where he funcion q{) is rel-vlued nd loclly

More information

Calculation method of flux measurements by static chambers

Calculation method of flux measurements by static chambers lculion mehod of flux mesuremens by sic chmbers P.S. Kroon Presened he NiroEurope Workshop, 15h - 17h December 28, openhgen, Denmrk EN-L--9-11 December 28 lculion mehod of flux mesuremens by sic chmbers

More information

RESPONSE UNDER A GENERAL PERIODIC FORCE. When the external force F(t) is periodic with periodτ = 2π

RESPONSE UNDER A GENERAL PERIODIC FORCE. When the external force F(t) is periodic with periodτ = 2π RESPONSE UNDER A GENERAL PERIODIC FORCE When he exernl force F() is periodic wih periodτ / ω,i cn be expnded in Fourier series F( ) o α ω α b ω () where τ F( ) ω d, τ,,,... () nd b τ F( ) ω d, τ,,... (3)

More information

Think of the Relationship Between Time and Space Again

Think of the Relationship Between Time and Space Again Repor nd Opinion, 1(3),009 hp://wwwsciencepubne sciencepub@gmilcom Think of he Relionship Beween Time nd Spce Agin Yng F-cheng Compny of Ruid Cenre in Xinjing 15 Hongxing Sree, Klmyi, Xingjing 834000,

More information

Hermite-Hadamard-Fejér type inequalities for convex functions via fractional integrals

Hermite-Hadamard-Fejér type inequalities for convex functions via fractional integrals Sud. Univ. Beş-Bolyi Mh. 6(5, No. 3, 355 366 Hermie-Hdmrd-Fejér ype inequliies for convex funcions vi frcionl inegrls İmd İşcn Asrc. In his pper, firsly we hve eslished Hermie Hdmrd-Fejér inequliy for

More information

A Robust DOA Estimation Based on Sigmoid Transform in Alpha Stable Noise Environment

A Robust DOA Estimation Based on Sigmoid Transform in Alpha Stable Noise Environment MAEC Web of Conferences 73 0006 (08) hps://doi.org/0.05/mecconf/08730006 MIMA 08 A obus DOA Esimion Bsed on igmoid rnsform in Alph ble oise Environmen Li Li * Mingyn e Informion Engineering College Dlin

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

Unified Framework for Developing Testing Effort Dependent Software Reliability Growth Models

Unified Framework for Developing Testing Effort Dependent Software Reliability Growth Models P. K. Kpur, Omr Shnwi, Anu G. Aggrwl, Rvi Kumr Unified Frmework for Developing Tesing Effor Dependen Sofwre Relibiliy Growh Models P.K. KAPUR 1, OMAR SHATNAI, ANU G. AGGARAL 1, RAVI KUMAR 1 1 Deprmen of

More information

HUI-HSIUNG KUO, ANUWAT SAE-TANG, AND BENEDYKT SZOZDA

HUI-HSIUNG KUO, ANUWAT SAE-TANG, AND BENEDYKT SZOZDA Communicions on Sochsic Anlysis Vol 6, No 4 2012 603-614 Serils Publicions wwwserilspublicionscom THE ITÔ FORMULA FOR A NEW STOCHASTIC INTEGRAL HUI-HSIUNG KUO, ANUWAT SAE-TANG, AND BENEDYKT SZOZDA Absrc

More information

Tax Audit and Vertical Externalities

Tax Audit and Vertical Externalities T Audi nd Vericl Eernliies Hidey Ko Misuyoshi Yngihr Ngoy Keizi Universiy Ngoy Universiy 1. Inroducion The vericl fiscl eernliies rise when he differen levels of governmens, such s he federl nd se governmens,

More information

P441 Analytical Mechanics - I. Coupled Oscillators. c Alex R. Dzierba

P441 Analytical Mechanics - I. Coupled Oscillators. c Alex R. Dzierba Lecure 3 Mondy - Deceber 5, 005 Wrien or ls upded: Deceber 3, 005 P44 Anlyicl Mechnics - I oupled Oscillors c Alex R. Dzierb oupled oscillors - rix echnique In Figure we show n exple of wo coupled oscillors,

More information

T-Match: Matching Techniques For Driving Yagi-Uda Antennas: T-Match. 2a s. Z in. (Sections 9.5 & 9.7 of Balanis)

T-Match: Matching Techniques For Driving Yagi-Uda Antennas: T-Match. 2a s. Z in. (Sections 9.5 & 9.7 of Balanis) 3/0/018 _mch.doc Pge 1 of 6 T-Mch: Mching Techniques For Driving Ygi-Ud Anenns: T-Mch (Secions 9.5 & 9.7 of Blnis) l s l / l / in The T-Mch is shun-mching echnique h cn be used o feed he driven elemen

More information

Volatility Forecasting with High Frequency Data

Volatility Forecasting with High Frequency Data Voliliy Forecsing wi Hig Frequency D oungjun Jng Deprmen of Economics Snford Universiy Snford, CA 945 jyjglory@gmil.com Advisor: Professor Peer Hnsen y 7, 7 ABSRAC e dily voliliy is ypiclly unobserved

More information

( ) ( ) ( ) ( ) ( ) ( y )

( ) ( ) ( ) ( ) ( ) ( y ) 8. Lengh of Plne Curve The mos fmous heorem in ll of mhemics is he Pyhgoren Theorem. I s formulion s he disnce formul is used o find he lenghs of line segmens in he coordine plne. In his secion you ll

More information

Endogenous Formation of Limit Order Books: Dynamics Between Trades.

Endogenous Formation of Limit Order Books: Dynamics Between Trades. Endogenous Formion of Limi Order Books: Dynmics Beween Trdes. Romn Gyduk nd Sergey Ndochiy Curren version: June 9, 7 Originl version: My 6, 6 Absrc In his work, we presen coninuous-ime lrge-populion gme

More information

A 1.3 m 2.5 m 2.8 m. x = m m = 8400 m. y = 4900 m 3200 m = 1700 m

A 1.3 m 2.5 m 2.8 m. x = m m = 8400 m. y = 4900 m 3200 m = 1700 m PHYS : Soluions o Chper 3 Home Work. SSM REASONING The displcemen is ecor drwn from he iniil posiion o he finl posiion. The mgniude of he displcemen is he shores disnce beween he posiions. Noe h i is onl

More information

Factorized Decision Forecasting via Combining Value-based and Reward-based Estimation

Factorized Decision Forecasting via Combining Value-based and Reward-based Estimation Fcorized Decision Forecsing vi Combining Vlue-bsed nd Rewrd-bsed Esimion Brin D. Ziebr Crnegie Mellon Universiy Pisburgh, PA 15213 bziebr@cs.cmu.edu Absrc A powerful recen perspecive for predicing sequenil

More information

Affine term structure models

Affine term structure models Affine erm srucure models A. Inro o Gaussian affine erm srucure models B. Esimaion by minimum chi square (Hamilon and Wu) C. Esimaion by OLS (Adrian, Moench, and Crump) D. Dynamic Nelson-Siegel model (Chrisensen,

More information

MTH 146 Class 11 Notes

MTH 146 Class 11 Notes 8.- Are of Surfce of Revoluion MTH 6 Clss Noes Suppose we wish o revolve curve C round n is nd find he surfce re of he resuling solid. Suppose f( ) is nonnegive funcion wih coninuous firs derivive on he

More information

Has the Business Cycle Changed? Evidence and Explanations. Appendix

Has the Business Cycle Changed? Evidence and Explanations. Appendix Has he Business Ccle Changed? Evidence and Explanaions Appendix Augus 2003 James H. Sock Deparmen of Economics, Harvard Universi and he Naional Bureau of Economic Research and Mark W. Wason* Woodrow Wilson

More information

RELATIONSHIP BETWEEN DISTRIBUTION AND GROWTH

RELATIONSHIP BETWEEN DISTRIBUTION AND GROWTH A POST-KEYNESIAN MODEL FOR ANALYZING THE RELATIONSHIP BETWEEN DISTRIBUTION AND GROWTH Engelber Sockhmmer* Özlem Onrn** prepred for he Conference on "Old nd New Growh Theories: An Assessmen" Ocober 5-7,

More information

Reinforcement Learning. Markov Decision Processes

Reinforcement Learning. Markov Decision Processes einforcemen Lerning Mrkov Decision rocesses Mnfred Huber 2014 1 equenil Decision Mking N-rmed bi problems re no good wy o model sequenil decision problem Only dels wih sic decision sequences Could be miiged

More information

A Dynamic Model of Economic Fluctuations

A Dynamic Model of Economic Fluctuations CHAPTER 15 A Dynamic Model of Economic Flucuaions Modified for ECON 2204 by Bob Murphy 2016 Worh Publishers, all righs reserved IN THIS CHAPTER, OU WILL LEARN: how o incorporae dynamics ino he AD-AS model

More information

Solutions for Nonlinear Partial Differential Equations By Tan-Cot Method

Solutions for Nonlinear Partial Differential Equations By Tan-Cot Method IOSR Journl of Mhemics (IOSR-JM) e-issn: 78-578. Volume 5, Issue 3 (Jn. - Feb. 13), PP 6-11 Soluions for Nonliner Pril Differenil Equions By Tn-Co Mehod Mhmood Jwd Abdul Rsool Abu Al-Sheer Al -Rfidin Universiy

More information

Thermal neutron self-shielding factor in foils: a universal curve

Thermal neutron self-shielding factor in foils: a universal curve Proceedings of he Inernionl Conference on Reserch Recor Uilizion, Sfey, Decommissioning, Fuel nd Wse Mngemen (Snigo, Chile, -4 Nov.3) Pper IAEA-CN-/, IAEA Proceedings Series, Vienn, 5 Therml neuron self-shielding

More information

Variable Growth Impacts on Optimal Market Timing in All-Out Production Systems

Variable Growth Impacts on Optimal Market Timing in All-Out Production Systems Vrible Growh Impcs on Opiml Mrke Timing in ll-ou Producion Sysems Jy R. Prsons, Dn L. Hog,. Mrshll Frsier, nd Sephen R. Koonz 1 bsrc This pper ddresses he economic impcs of growh vribiliy on mrke iming

More information

Generalized Least Squares

Generalized Least Squares Generalized Leas Squares Augus 006 1 Modified Model Original assumpions: 1 Specificaion: y = Xβ + ε (1) Eε =0 3 EX 0 ε =0 4 Eεε 0 = σ I In his secion, we consider relaxing assumpion (4) Insead, assume

More information

Credit Ratings and Corporate Investment: UK Evidence

Credit Ratings and Corporate Investment: UK Evidence Credi Rings nd Corpore Invesmen: UK Evidence Credi Rings nd Corpore Invesmen: UK Evidence Hong Bo Deprmen of Finncil & Mngemen Sudies SOAS, Universiy of London, UK Rober Lensink Deprmen of Finnce Universiy

More information

ENGR 1990 Engineering Mathematics The Integral of a Function as a Function

ENGR 1990 Engineering Mathematics The Integral of a Function as a Function ENGR 1990 Engineering Mhemics The Inegrl of Funcion s Funcion Previously, we lerned how o esime he inegrl of funcion f( ) over some inervl y dding he res of finie se of rpezoids h represen he re under

More information

A Multi-Cycle Two-Factor Model of Asset Replacement

A Multi-Cycle Two-Factor Model of Asset Replacement A Muli-ycle Two-Fcor Model of Asse Replcemen Februry 009 João Z. Oliveir Deprmen of Mngemen nd Economics Universiy of Mdeir mpus d Pened 9050-590 Funchl, Porugl Tel: (35) 97 67 68 Fx: (35) 9 705 040 Emil:

More information