ASYMPTOTIC EQUIVALENCE OF NONPARAMETRIC REGRESSION AND WHITE NOISE. BY LAWRENCE D. BROWN 1 AND MARK G. LOW 2 University of Pennsylvania

Size: px
Start display at page:

Download "ASYMPTOTIC EQUIVALENCE OF NONPARAMETRIC REGRESSION AND WHITE NOISE. BY LAWRENCE D. BROWN 1 AND MARK G. LOW 2 University of Pennsylvania"

Transcription

1 The Aals of Sascs 996, Vol., No. 6, ASYMPTOTIC EQUIVALENCE OF NONPARAMETRIC REGRESSION AND WITE NOISE BY LAWRENCE D. BROWN AND MARK G. LOW Uversy of Pesylvaa The prcpal resul s ha, uder codos, o ay oparamerc regresso problem here correspods a asympocally equvale sequece of whe ose wh drf problems, ad coversely. Ths asympoc equvalece s a global ad uform sese. Ay ormalzed rsk fuco aaable oe problem s asympocally aaable he oher, wh he dfferece ormalzed rsks covergg o zero uformly over he ere parameer space. The resuls are cosrucve. A recpe s provded for producg hese asympocally equvale procedures. Some mplcaos ad geeralzaos of he prcpal resul are also dscussed. Iroduco. The prcpal resul of hs paper s ha o ay oparamerc regresso problem here correspods a whe ose wh drf problem whch s asympocally equvale. The mpac of hs asympoc equvalece s ha ay asympoc soluo o oe of hese problems wll auomacally yeld a correspodg soluo o he oher. I addo, here s a explc recpe for hs correspodece. For example, he opmal raes of covergece wll be equal as wll suably ormalzed local ad global asympoc rsks, ad kowledge of a mmax procedure or of a lear mmax procedure oe problem auomacally yelds he correspodg procedure he oher ad so forh. I parcular, may classcal fucoal esmao problems whch have prevously bee reaed separaely fall o hs framework. These problems clude:. Esmag he whole fuco, cosdered he whe-ose model by Psker 980. ad he regresso coex by Nussbaum 985. ad Speckma See also Dooho, Lu ad MacGbbo 990. ad Golubev ad Nussbaum Esmag a po fucoal such as f x. 0. For whe ose, see Ibragmov ad asmsk 98. ad for regresso, see Ibragmov ad asmsk 98.. See also Dooho ad Lu 99., pages 677ff ad 688 ad Dooho ad Low 99.. Receved Jauary 99; revsed Aprl 993. Suppored par by NSF Gras DMS-9-078, 93-08, Suppored par by NSF Gra DMS AMS 99 subjec classfcaos. Prmary 6G07; secodary 6G0, 6M05. Key words ad phrases. Rsk equvalece, local asympoc mmaxy, lear esmaors. 38

2 NONPARAMETRIC REGRESSION AND WITE NOISE Esmag a olear fucoal such as f x. dx, reaed for whe ose Fa 99b. ad earler refereces ced here ad for boh whe ose ad regresso Dooho ad Nussbaum Esmag he whole fuco based o drec observaos as Fa 99a.. The equvalece heory also covers may adapve suaos. If here s a parcular sequece of esmaors whch s asympocally mmax over a colleco of parameer spaces he whe-ose case, he here s a correspodg sequece based o he regresso model whch s also mmax over each of hese parameer spaces. Such a sequece of adapve esmaors for esmag he whole fuco has bee foud by Efromovch ad Psker 98. he whe-ose case ad by Golubev 987. for oparamerc regresso. See also Golubev 99.. For he problem of esmag a lear fucoal boh regresso ad whe ose, see Lepsk 99.. all ad Johsoe 99. dscuss examples of esmag opmal, possbly radom, badwdhs he regresso ad whe-ose coexs. These problems are also covered by he geeral equvalece heory developed hs paper. See Remark.5. The equvalece heory ca provde a addoal echcal advaage. Some proofs, for example hose volvg raes of covergece, may be much smpler he whe-ose model. Thus oe may use he whe-ose model o fgure ou he opmal rae va homogeey ad he he same rae holds oparamerc regresso. See, for example, Seco 7 of Dooho ad Low 99.. Aalogous equvalece resuls should be vald for some oher oparamerc problems. Ideed, Nussbaum 993. has very recely proved oe such resul for oparamerc desy esmao. The frs par of he paper coas ecessary backgroud. Ths cludes descrpos of he oparamerc regresso ad whe-ose problems, he defo of asympoc equvalece ad a dscusso of some of s geeral cosequeces. The secod par of he paper coas he ma equvalece heorems. Two cases are reaed separaely. I oe case, he depede varables are deermscally fxed; he oher, hey are a radom sample from a specfed dsrbuo. Par. Backgroud.. Noparamerc regresso. The oparamerc regresso model o be reaed hs paper s as follows: Le I be a possbly fe erval. Le f.: I ad.: I 0,. be wo measurable fucos ad le : 0, be a creasg c.d.f. The varables X, Y.,,...,, are observed. I he deermsc X vara he depede varables are gve by a deermsc scheme whch wll be gve by.. x.,,...,,

3 386 L. D. BROWN AND M. G. LOW excep where oherwse oed. I he radom X vara he X are depe de radom varables.. X..d.,...,. I eher case he codoal dsrbuo of Y gve X s descrbed va.3. Y f x. x., N 0,. d.,,...,. The parameer space cosss of a possbly large se of choces of f. The c.d.f. ad he fuco are assumed fxed ad kow pror o expermeao. ere are some examples whch are subjec o laer resuls of hs paper. EXAMPLE.. s gve erms of a Lpshz codo as...., B ½ Ý. f : f x f x f x! B ad sup f x. B f, 5 xi. f : f x. f x. B f 0., B. holds for all x, x I. Laer we shall requre, as explaed Remark.7. EXAMPLE.. s gve by a Sobolev ype codo such as. 5 xi.. f : f d B ad sup f x B, B ½ for,..., where f. deoes he h dervave of f, assumed o exs he sese ha f. s absoluely couous. Noe ha whe I d, he,. B., B.. Whe ose. Assume wh o loss of geeraly ha 0 I. Le B : I deoe Browa moo o I codoal o B0 0, for example, B N 0,. ad B, 0, s depede of B, 0. Fx,.... Le Z deoe he Gaussa process whose whe-ose verso s repre- seed as dz. d. db '. The parameer space hs model s, as before, he se of possble mea fucos. Cosequely, he sascal whe-ose problem for gve s o observe he process Z defed above for some. I hs way a sequece of problems s defed for,,..., each of whch has he same parameer space.

4 NONPARAMETRIC REGRESSION AND WITE NOISE 387 Suppose Y : J ad Z : J are wo Gaussa processes, lke hose above, wh respecve mea fucos. ad. ad wh decal varace fucos. Le g Y ad gz deoe her respecve probably deses wh respec o ay domag measure. Le. L g. g. d.. Y Z Where covee we also use eher he oao L Y, Z. or L g, g.. Sadard calculaos yeld Y Z. L D. where Cosequely, L O D. D.... d. REMARK.. Le Z be as above. Assume s absoluely couous wh dd h ad assume h 0 a.e.. o I. Defe he Gaussa process V Z, 0,. The V has mea. ad varace fuco. gve by ,.. h. h. ece here s o sgfca loss of geeraly assumg I 0, ad s uform o I, alhough o do so wll affec he defo of, he maer suggesed by Sascal equvalece. Cosder wo sascal problems, P. ad.. P, wh sample spaces X,, ad suable -felds., respecvely, bu wh he same parameer space. Deoe he respecve famles of dsrbu-. os by G :. The followg paragraph descrbes Le Cam s merc for he dsace bewee wo such expermes see, e.g., Le Cam 986. or Le Cam ad Yag Le A be ay measurable. aco space ad le L: A 0,. deoe a. loss fuco. Le L sup L, a :, a A. wll be he geerc symbol for a radomzed. decso procedure he h problem. The rsk from usg procedure. whe L s he loss fuco ad s he rue value.. of he parameer s deoed by R, L,.. Le Cam s merc s 3.. P., P max f sup sup sup R, L, R, L,,.. L: L.... f sup sup sup R, L, R, L,... L: L

5 388 L. D. BROWN AND M. G. LOW. Thus, f, hs meas ha for every procedure problem here s a procedure j. problem j, j, wh rsk dfferg by a mos, uformly over all L such ha L ad.. Two sequeces of.. problems P :,... ad P :,... are asympocally equvale.. f P, P. 0 as. I hs case, for ay sequece of procedures. problems P.,,..., here s a sequece. problems P. for whch.... lm sup sup R, L, R L,.. 0. L: L Such sequeces of procedures are sad o be asympocally equvale. I our proofs of sascal equvalece he key sep s o arrage maers so ha X. X.. The defe L P, P. sup g x. g x. dx., where domaes G. ad g. dg. d,,. The followg wellkow fac ca he be used o esablsh asympoc equvalece. TEOREM R, L,. R, L,. L P, P L. Cosequely, 3.. P., P.. L P., P... Also, PROOF. P., P.. 0 f L P., P s jus a resaeme of he sadard equaly... h x g x g x dx ž /... sup h x. g x. g x. dx.. Two oher echques are also used repeaedly: oe s reduco by suffcecy; he oher volves relaos bewee he L orm ad he ellger merc. The use of suffcecy s based o he followg. LEMMA 3.. Le X ad s -feld be a Polsh space wh s assocaed Borel feld. Le P. deoe a experme wh sample space X. Le S: X Y. be a suffce sasc ad le P deoe he experme whch Y S X... s observed. The P, P. 0. A Polsh space s oe whch s locally compac, merzable ad secod couable. All he measurable spaces ecouered hs paper possess hese properes.. PROOF. The lemma follows from he fac ha uder he hypohess here exss a measurable map dx y. such ha BS x.. G dx. G B.. See Le Cam 986..

6 NONPARAMETRIC REGRESSION AND WITE NOISE The ellger merc G, G s defed by I provdes a boud for L.... G, G g x g x dx. sce 3.5. L G., G.. G., G... Aoher useful fac s ha f he G. are produc measures, G. Ł m jgj., he.. j j m G, G G, G. Ł j Le Cam 986. Fally, drec calculao yelds N,, N,. exp. Par. Equvalece resuls.. Deermsc X. Le I,,. Le 0 be a gve absoluely couous fuco o I such ha.. l. C, I. for some C. Suppose.. sup f. : I, f B ad also.5, below. Assume s absoluely couous o I ad.3.. h. 0 a.e. o I. Wh x as.. defe he sep fuco.. f. ½ f x.,,,...,, f x.,, where.. The depedece of o s suppressed for cove- ece.. Assume..5. lm sup f. f. h. d 0. f REMARK.. Equao.5. s a uform smoohess codo o f. I s sasfed a wde varey of examples. I parcular s sasfed Examples. ad. whe. See Remark.7 for dscusso of he case..

7 390 L. D. BROWN AND M. G. LOW For example, o verfy.5. Example. for he case, I 0,, uform ad we oe ha f. f. B., where.. ece. Ý 0. B ž / 0 f f d B d sce. ece.5 s sasfed. B d 0 TEOREM.. Uder assumpos...5. he deermsc X oparamerc regresso model s asympocally equvale o he whe-ose model. havg. f.,.. h.. PROOF. Le Z be he whe-ose model descrbed by. dz f. d db. ' Noe ha sup.: If.: I because of.. ad.3.. ece.5.,. ad Theorem 3. show ha Defe. Z, Z 0. x. dz.6. K ad S K.. d.. The varables S :,..., are suffce for Z : I. ece. Z, S 0 by Lemma 3.. The varables S,,...,, are depede wh K d V S x.,. f E S. K d..7. h. ' x. f x. d. x. f x.,.

8 NONPARAMETRIC REGRESSION AND WITE NOISE 39 where. The exsece of follows from he mea value heorem sce h. d. Assumpo.. he yelds E S f x. O. uformly f, ad. Noe also ha he O erm s zero f, a cosa. The above, ogeher wh 3.6. ad 3.7., yelds.8.. S, x, Y ž Ý.. / O f x. uformly over f. B b a. O sup. :,..., o by. ad.3. ece lm S, x, Y. 0 by 3.5. ad Theorem. I follows from he precedg facs ha lm Z, x, Y. 0. TECNICAL NOTE. The codos...3. are clearly sroger ha ecessary. They are used order o yeld he cocluso o. u-. formly over f whch appears.8. For example, f, a cosa, I 0, ad h. o 0,, he.3. holds ad we may also drop codos.. ad... The above heorem yelds a prescrpo for producg, from a sequece of procedures oe problem, a asympocally equvale sequece he oher. The followg corollary gves a precse recpe. The corollary apples o eher oradomzed or radomzed procedures. COROLLARY.. Le be a sequece of procedures he regresso model of Theorem.. Defe he correspodg whe-ose problem by dz..9. Z. S. where S K as.6.. The s asympocally equvale o. Coversely, suppose s a gve sequece of procedures he whe-ose problem. The s a asympocally equvale sequece he asympo- cally equvale regresso problem, where s he radomzed procedure descrbed for ay measurable A A as follows: dz.. ž /.0 A s E A Z K d s,,...,. PROOF. The corollary s mplc he proof of Theorem 5.. Noe ha.0 s well defed sce s depede of he drf parameer f. of he Z process because for each, S s suffce for Z.

9 39 L. D. BROWN AND M. G. LOW The cosruco.0. s o erely felcous for wo reasos:. wll, geeral, be radomzed eve whe s o; he codoal expeca- o may be hard o evaluae. The followg wo remarks show ha hese dffcules ca somemes be parally allevaed. REMARK.. Jese s equaly ca somemes be used o mprove.0.. ere s a precse saeme. Suppose each A s a closed covex subse of a separable Baach space, ad suppose each L,. s a covex lower sem- couous fuco sasfyg lm a L, a.. Defe as he oradomzed procedure akg he values. s a da s. The s asympocally a leas as good as. Eve whe L s o covex may be ha.. sup R, L B,. R, L B,. 0. whch case ad are asympocally equvale sequeces for L. See Remark.3 for oe such suao. A real-valued. lear esmaor he whe-ose problem s a oradomzed esmaor whch ca be expressed he form.3. Z. dz, wh. d. I follows ha Z. has a ormal dsrbuo wh mea.. d ad varace.. d. A lear esmaor he regresso problem s, smlarly, oe whch ca be expressed he form. s s. Ý As above, such esmaors are ormally dsrbued. REMARK.3. The equvalece rasformao s especally covee for lear esmaors. I order o demosrae hs, s easer o wre.. a alerae form. As before, for coveece, specalze o he case I 0,, s uform, ad also assume. Le e. deoe he ormalzed learzed cumulave sums of s ; ha s, ½ Ý m m. 5 k e s s s,,,...,.

10 NONPARAMETRIC REGRESSION AND WITE NOISE 393 The he geeral lear esmaor. ca be rewre he form.5 s de. 0 Noe ha dz. e E Z K d s,,...,. ere, ad K. ece, suppose a lear esmaor.3. s gve he whe-ose problem. The he esmaor of.. s gve by.5. wh... The precedg resuls apply o regresso problems o a bouded erval. They ca be exeded whou much addoal complcao o also apply o ubouded ervals f a somewha dffere scheme for deermg he X values s roduced, as follows: Le I a, b. ad I,, a b ad le be a creasg c.d.f. o,. Choose he depede varables accord- g o.6. x.. The cosas whch also deped o. are defed as before, bu wh place of, ad f s defed by.. wh he addoal coveo f. 0,,. The assumpo.5 s he replaced by.7 lm sup h f f. d 0. f ere s a formal saeme of he equvalece resul. Is proof volves oly mor modfcaos of he proof of Theorem., ad wll be omed. COROLLARY.. Defe x as above. Assume.... are sasfed o, ad sasfes.3. Assume.7 s sasfed ad eher, a cosa, or max o. ere., 0 ad. As usual, he depedece of o s suppressed he oao. The he deermsc X oparamerc regresso model s asympocally equvale o he whe-ose model. havg. f. ad cremeal varace fuco depedg o ad gve by.. h.. REMARK.. There s a addoal queso whch arses he preced- g suao: Suppose a b ad wh h. Le Z be he whe ose wh drf f.. ad wh varace fuco. h. Is rue ha x, Y, Z. 0?

11 39 L. D. BROWN AND M. G. LOW A affrmave aswer o he precedg queso appears o requre some codo o he covergece of o. I some examples s o hard o supply such a codo. Cosder, for example, he suao Example. wh. Assume for smplcy of exposo ha. Make he oher assumpos Theorem. ad assume also ha.8. sup.. o '. as. The he aswer s affrmave. To esablsh he precedg clam, le x, Y deoe he deermsc X oparamerc regresso based o. The x, Y, x, Y. Ý f ž f B 0. ž // ž ž //. The apply Theorem. o see ha x, Y, Z 0. REMARK.5. The precedg mehodology ca be used whe he loss fucos also deped o he observaos. Thus, suppose he loss he. regresso problem s L : A x, y. 0, B ad he whe-ose. problem s L : A Z 0, B. For smplcy assume, a cosa. Defe T from Z he same maer as S was defed from Z see.6.. Assume he wo loss fucos asympocally agree he sese ha for each f, E L f, a, Z L f, a, x, T 0 f uformly a A. The proof of Theorem. ca he easly be adaped o prove ha correspodg procedures he wo problems have asympocally equal rsk fucos uder he respecve losses L. ad L.. The precedg observao ca be used o prove equvalece he problem of opmal badwdh seleco as formulaed all ad Johsoe 99.. The codo.9. s relavely sraghforward o check her coex. The resulg cocluso s ha her asympoc resul, oce proved he whe-ose seg, s he also vald he oparamerc regresso seg. REMARK.6. Lookg a Example. he case where shows wha ca happe whe he key regulary codo.5. fals. Le I 0,, h ad. I he whe-ose problem ' Z d N 0, 0

12 NONPARAMETRIC REGRESSION AND WITE NOISE 395 dsrbuo uformly over. I fac he lef sde s exacly N 0,.. owever, he regresso problem of Seco here cao exs a esmaor d. such ha.,. '..0 d x, Y f d N 0, 0 uformly over f,... By vrue of Remark 3., hs shows ha he wo problems are o asympocally equvale. To verfy he mpossbly of.0., le ', where. s defed by.. m j.: j,...,. The f. 0 ad f.. are. boh, ad x, Y.,. has he decal dsrbuo boh cases. ' owever,. d ' 3. ece valdy of.0. for f. 0 0 mples s falure for f... Alhough he wo sequeces of problems are o asympocally equvale he srog sese of Theorem., everheless may specal cases hey are asympocally equally useful. For example, hey wll have he same local asympoc mmax rsks for esmag f x. 0 uder squared error loss or for esmag f. uder egraed squared error loss. 5. Radom X. Resuls aalogous o hose of he precedg seco ca be derved for he radom X oparamerc regresso problem, as defed... I hs case he oparamerc regresso problem s asympocally equvale o a whe-ose model whch he drf depeds o he observed values of X, as well as he ukow parameers. To descrbe hs whe-ose model o I,,, le x.,..., x deoe he ordered values of X ad le x0., x. ad le 5.. x... x x... f x. x.,,...,. Now defe ad defe f by. wh place of. Le h. deoe he lef-had dervave of a. I place of.5. assume lm prob sup f. f. h.. d 0. f REMARK 5.. Assumpo 5.. s usually o much harder o verfy ha s.5.. For example, f I 0,, s he uform dsrbuo,. ad Example., he 5.. follows sce. ž../ f f h d f v f v dv

13 396 L. D. BROWN AND M. G. LOW ad ½ ž / 5 j. j.. E f v f v B E x x 0. uformly v,, where j m : x v. I he precedg suao defe.. h. ad. X, dz f. d db. ' X,. Noe ha he dsrbuo of Z depeds o he acllary observaos X hrough he local varace fuco. TEOREM 5.. Assume, f, sasfy...3. ad 5... The he radom X oparamerc regresso model s asympocally equvale o he whe-ose model PROOF. Aalogously o Theorem., le. X,. dz f. d db. ' The remader of he proof exacly follows he paer of proof of Theorem. wh Z place of Z ad f place of f. Also.8. eeds o be modfed slghly o beg as ž /. ž Ý / 5.. X, S, X, Y O E f x.. ad so forh, sce x ad are radom. A alerave proof could be based drecly o Corollary... REMARK 5.. A cospcuous feaure of he precedg resul s ha kowledge of s o requred o cosruc Z X,. or o carry hrough he proof of he heorem. Thus, suppose s oly assumed ha, where.. lm prob sup sup f f h d 0 f ad also ha eher or.3. s modfed o requre he exsece of a h sasfyg 0 f h. : h h. 0 a.e. o I. The he equvalece assero of Theorem 5. remas vald. Noe also ha he cosruco Corollary. of equvale procedures ogeher wh he coes of Remarks.. ca easly be carred over o he curre suao. 0

14 NONPARAMETRIC REGRESSION AND WITE NOISE 397 REMARK 5.3. Suppose s kow. A ssue whch he arses s wheher hs kowledge ca serve place of observao of he acllary sascs X,..., X. More precsely, le Xj Xj deoe he ordered values of X :,..., ad le Y,..., Y deoe he correspodg values of Y j j. The gve kowledge of, s he experme Y :,..., j asympo- cally equvale o X, Y.:,...,? Because of Theorem 5., hs s X,. equvale o askg wheher he experme Z : I s asympocally X,. equvale o, Z.: I. The geeral aswer o hs queso s No. Suppose I 0,,. for I ad f: f. c, c. The he experme Z X,. he X,. UMVU ad mmax esmaor s dz, whch has varace ž / 5.5 d Var d 3. Meawhle, for he experme, Z X,. he UMVU ad mmax esma- X,. or s dz, whch has varace. Ths shows he wo exper- mes are o asympocally equvale. O he oher had, for may purposes he wo expermes are equally useful asympocally. Roughly speakg hs happes whe he rae of covergece of he esmaor used s slower ha he classcal ' rae. eurscally, wha happes such problems s ha he mprecso ro- s represeed by a erm aalogous o he secod duced by o observg oe o he lef of 5.5., whch s of order. Whe he esmaor coverges ' a slower ha rae, so ha s varace coverges more slowly ha, hs secod erm s asympocally eglgble. Such behavor ca be deduced examples lke hose ced Dooho ad Lu 99. ad Low 99.. The auhors hak Luce Le Cam for helpful co- Ackowledgme. versaos. REFERENCE DONOO, D. L. ad LIU, R. C Geomerzg raes of covergece III. A. Sas DONOO, D. L., LIU, R. C. ad MACGIBBON, B Mmax rsk for hyperrecagles. A. Sas DONOO, D. L. ad LOW, M. C Reormalzao expoes ad opmal powse raes of covergece. A. Sas DONOO, D. L. ad NUSSBAUM, M Mmax quadrac esmao of a quadrac fucoal. J. Complexy EFROIMOVIC, S. Y. ad PINSKER, M. S A learg algorhm for oparamerc flerg. Auoma. Telemekh I Russa.. FAN, J. 99a.. O he opmal raes of covergece for oparamerc decovoluo problems. A. Sas FAN, J. 99b.. O he esmao of quadrac fucos. A. Sas GOLUBEV, G. K Adapve asympocally mmax esmaes of smooh sgals. Problemy Peredach Iformas I Russa.. GOLUBEV, G. K LAN oparamerc fucoal esmao problems ad lower bouds for quadrac rsk. Teorya Veroyaos. Prmee I Russa..

15 398 L. D. BROWN AND M. G. LOW GOLBEV, K. ad NUSSBAUM, M A rsk boud Sobolev class regresso. A. Sas ALL, P. ad JONSTONE, I. M Emprcal fucoals ad effce smoohg parameer seleco. J. Roy. Sas. Soc. Ser. B IBRAGIMOV, I. A. ad ASMINSKII, R. Z Bouds for he rsk of oparamerc regresso esmaes. Theory Probab. Appl Eglsh raslao.. IBRAGIMOV, I. A. ad ASMINSKII, R. Z O oparamerc esmao of he value of a lear fucoal Gaussa whe ose. Theory Probab. Appl Eglsh raslao.. LE CAM, L Asympoc Mehods Sascal Decso Theory. Sprger, New York. LE CAM, L. ad YANG, G. L Asympocs Sascs: Some Basc Coceps. Sprger, New York. LEPSKII, O. V Asympocally mmax adapve esmao I. Upper bouds. Opmally adapve esmaes. Theory Probab. Appl LOW, M. G Reormalzao ad whe ose approxmao for oparamerc fucoal esmao problems. A. Sas NUSSBAUM, M Sple smoohg regresso models ad asympoc effcecy L. A. Sas NUSSBAUM, M Asympoc equvalece of desy esmao ad whe ose. A. Sas PINSKER, M. S Opmal flerg of square egrable sgals Gaussa whe ose. Problems Iform. Trasmsso Eglsh raslao.. SPECKMAN, P Sple smoohg ad opmal raes of covergece oparamerc regresso models. A. Sas DEPARTMENT OF STATISTICS UNIVERSITY OF PENNSYLVANIA 360 LOCUST WALK PILADELPIA, PENNSYLVANIA 90 lbrow@compsa.wharo.upe.edu lowm@compsa.wharo.upe.edu

The Poisson Process Properties of the Poisson Process

The Poisson Process Properties of the Poisson Process Posso Processes Summary The Posso Process Properes of he Posso Process Ierarrval mes Memoryless propery ad he resdual lfeme paradox Superposo of Posso processes Radom seleco of Posso Pos Bulk Arrvals ad

More information

Asymptotic Equivalence of Nonparametric Regression and White Noise

Asymptotic Equivalence of Nonparametric Regression and White Noise Uversty of Pesylvaa ScholarlyCommos Statstcs Papers Wharto Faculty Research 1996 Asymptotc Equvalece of Noparametrc Regresso ad Whte Nose Lawrece D. Brow Uversty of Pesylvaa Mark G. Low Uversty of Pesylvaa

More information

14. Poisson Processes

14. Poisson Processes 4. Posso Processes I Lecure 4 we roduced Posso arrvals as he lmg behavor of Bomal radom varables. Refer o Posso approxmao of Bomal radom varables. From he dscusso here see 4-6-4-8 Lecure 4 " arrvals occur

More information

Fundamentals of Speech Recognition Suggested Project The Hidden Markov Model

Fundamentals of Speech Recognition Suggested Project The Hidden Markov Model . Projec Iroduco Fudameals of Speech Recogo Suggesed Projec The Hdde Markov Model For hs projec, s proposed ha you desg ad mpleme a hdde Markov model (HMM) ha opmally maches he behavor of a se of rag sequeces

More information

Least Squares Fitting (LSQF) with a complicated function Theexampleswehavelookedatsofarhavebeenlinearintheparameters

Least Squares Fitting (LSQF) with a complicated function Theexampleswehavelookedatsofarhavebeenlinearintheparameters Leas Squares Fg LSQF wh a complcaed fuco Theeampleswehavelookedasofarhavebeelearheparameers ha we have bee rg o deerme e.g. slope, ercep. For he case where he fuco s lear he parameers we ca fd a aalc soluo

More information

Key words: Fractional difference equation, oscillatory solutions,

Key words: Fractional difference equation, oscillatory solutions, OSCILLATION PROPERTIES OF SOLUTIONS OF FRACTIONAL DIFFERENCE EQUATIONS Musafa BAYRAM * ad Ayd SECER * Deparme of Compuer Egeerg, Isabul Gelsm Uversy Deparme of Mahemacal Egeerg, Yldz Techcal Uversy * Correspodg

More information

Determination of Antoine Equation Parameters. December 4, 2012 PreFEED Corporation Yoshio Kumagae. Introduction

Determination of Antoine Equation Parameters. December 4, 2012 PreFEED Corporation Yoshio Kumagae. Introduction refeed Soluos for R&D o Desg Deermao of oe Equao arameers Soluos for R&D o Desg December 4, 0 refeed orporao Yosho Kumagae refeed Iroduco hyscal propery daa s exremely mpora for performg process desg ad

More information

The Linear Regression Of Weighted Segments

The Linear Regression Of Weighted Segments The Lear Regresso Of Weghed Segmes George Dael Maeescu Absrac. We proposed a regresso model where he depede varable s made o up of pos bu segmes. Ths suao correspods o he markes hroughou he da are observed

More information

As evident from the full-sample-model, we continue to assume that individual errors are identically and

As evident from the full-sample-model, we continue to assume that individual errors are identically and Maxmum Lkelhood smao Greee Ch.4; App. R scrp modsa, modsb If we feel safe makg assumpos o he sascal dsrbuo of he error erm, Maxmum Lkelhood smao (ML) s a aracve alerave o Leas Squares for lear regresso

More information

Asymptotic Behavior of Solutions of Nonlinear Delay Differential Equations With Impulse

Asymptotic Behavior of Solutions of Nonlinear Delay Differential Equations With Impulse P a g e Vol Issue7Ver,oveber Global Joural of Scece Froer Research Asypoc Behavor of Soluos of olear Delay Dffereal Equaos Wh Ipulse Zhag xog GJSFR Classfcao - F FOR 3 Absrac Ths paper sudes he asypoc

More information

(1) Cov(, ) E[( E( ))( E( ))]

(1) Cov(, ) E[( E( ))( E( ))] Impac of Auocorrelao o OLS Esmaes ECON 3033/Evas Cosder a smple bvarae me-seres model of he form: y 0 x The four key assumpos abou ε hs model are ) E(ε ) = E[ε x ]=0 ) Var(ε ) =Var(ε x ) = ) Cov(ε, ε )

More information

The Mean Residual Lifetime of (n k + 1)-out-of-n Systems in Discrete Setting

The Mean Residual Lifetime of (n k + 1)-out-of-n Systems in Discrete Setting Appled Mahemacs 4 5 466-477 Publshed Ole February 4 (hp//wwwscrporg/oural/am hp//dxdoorg/436/am45346 The Mea Resdual Lfeme of ( + -ou-of- Sysems Dscree Seg Maryam Torab Sahboom Deparme of Sascs Scece ad

More information

QR factorization. Let P 1, P 2, P n-1, be matrices such that Pn 1Pn 2... PPA

QR factorization. Let P 1, P 2, P n-1, be matrices such that Pn 1Pn 2... PPA QR facorzao Ay x real marx ca be wre as AQR, where Q s orhogoal ad R s upper ragular. To oba Q ad R, we use he Householder rasformao as follows: Le P, P, P -, be marces such ha P P... PPA ( R s upper ragular.

More information

Some Probability Inequalities for Quadratic Forms of Negatively Dependent Subgaussian Random Variables

Some Probability Inequalities for Quadratic Forms of Negatively Dependent Subgaussian Random Variables Joural of Sceces Islamc epublc of Ira 6(: 63-67 (005 Uvers of ehra ISSN 06-04 hp://scecesuacr Some Probabl Iequales for Quadrac Forms of Negavel Depede Subgaussa adom Varables M Am A ozorga ad H Zare 3

More information

Moments of Order Statistics from Nonidentically Distributed Three Parameters Beta typei and Erlang Truncated Exponential Variables

Moments of Order Statistics from Nonidentically Distributed Three Parameters Beta typei and Erlang Truncated Exponential Variables Joural of Mahemacs ad Sascs 6 (4): 442-448, 200 SSN 549-3644 200 Scece Publcaos Momes of Order Sascs from Nodecally Dsrbued Three Parameers Bea ype ad Erlag Trucaed Expoeal Varables A.A. Jamoom ad Z.A.

More information

Chapter 8. Simple Linear Regression

Chapter 8. Simple Linear Regression Chaper 8. Smple Lear Regresso Regresso aalyss: regresso aalyss s a sascal mehodology o esmae he relaoshp of a respose varable o a se of predcor varable. whe here s jus oe predcor varable, we wll use smple

More information

Quantum Mechanics II Lecture 11 Time-dependent perturbation theory. Time-dependent perturbation theory (degenerate or non-degenerate starting state)

Quantum Mechanics II Lecture 11 Time-dependent perturbation theory. Time-dependent perturbation theory (degenerate or non-degenerate starting state) Pro. O. B. Wrgh, Auum Quaum Mechacs II Lecure Tme-depede perurbao heory Tme-depede perurbao heory (degeerae or o-degeerae sarg sae) Cosder a sgle parcle whch, s uperurbed codo wh Hamloa H, ca exs a superposo

More information

4. THE DENSITY MATRIX

4. THE DENSITY MATRIX 4. THE DENSTY MATRX The desy marx or desy operaor s a alerae represeao of he sae of a quaum sysem for whch we have prevously used he wavefuco. Alhough descrbg a quaum sysem wh he desy marx s equvale o

More information

Continuous Time Markov Chains

Continuous Time Markov Chains Couous me Markov chas have seay sae probably soluos f a oly f hey are ergoc, us lke scree me Markov chas. Fg he seay sae probably vecor for a couous me Markov cha s o more ffcul ha s he scree me case,

More information

VARIATIONAL ITERATION METHOD FOR DELAY DIFFERENTIAL-ALGEBRAIC EQUATIONS. Hunan , China,

VARIATIONAL ITERATION METHOD FOR DELAY DIFFERENTIAL-ALGEBRAIC EQUATIONS. Hunan , China, Mahemacal ad Compuaoal Applcaos Vol. 5 No. 5 pp. 834-839. Assocao for Scefc Research VARIATIONAL ITERATION METHOD FOR DELAY DIFFERENTIAL-ALGEBRAIC EQUATIONS Hoglag Lu Aguo Xao Yogxag Zhao School of Mahemacs

More information

Solution. The straightforward approach is surprisingly difficult because one has to be careful about the limits.

Solution. The straightforward approach is surprisingly difficult because one has to be careful about the limits. ose ad Varably Homewor # (8), aswers Q: Power spera of some smple oses A Posso ose A Posso ose () s a sequee of dela-fuo pulses, eah ourrg depedely, a some rae r (More formally, s a sum of pulses of wdh

More information

Supplement Material for Inverse Probability Weighted Estimation of Local Average Treatment Effects: A Higher Order MSE Expansion

Supplement Material for Inverse Probability Weighted Estimation of Local Average Treatment Effects: A Higher Order MSE Expansion Suppleme Maeral for Iverse Probably Weged Esmao of Local Average Treame Effecs: A Hger Order MSE Expaso Sepe G. Doald Deparme of Ecoomcs Uversy of Texas a Aus Yu-C Hsu Isue of Ecoomcs Academa Sca Rober

More information

-distributed random variables consisting of n samples each. Determine the asymptotic confidence intervals for

-distributed random variables consisting of n samples each. Determine the asymptotic confidence intervals for Assgme Sepha Brumme Ocober 8h, 003 9 h semeser, 70544 PREFACE I 004, I ed o sped wo semesers o a sudy abroad as a posgraduae exchage sude a he Uversy of Techology Sydey, Ausrala. Each opporuy o ehace my

More information

Fault Tolerant Computing. Fault Tolerant Computing CS 530 Probabilistic methods: overview

Fault Tolerant Computing. Fault Tolerant Computing CS 530 Probabilistic methods: overview Probably 1/19/ CS 53 Probablsc mehods: overvew Yashwa K. Malaya Colorado Sae Uversy 1 Probablsc Mehods: Overvew Cocree umbers presece of uceray Probably Dsjo eves Sascal depedece Radom varables ad dsrbuos

More information

Linear Regression Linear Regression with Shrinkage

Linear Regression Linear Regression with Shrinkage Lear Regresso Lear Regresso h Shrkage Iroduco Regresso meas predcg a couous (usuall scalar oupu from a vecor of couous pus (feaures x. Example: Predcg vehcle fuel effcec (mpg from 8 arbues: Lear Regresso

More information

Least squares and motion. Nuno Vasconcelos ECE Department, UCSD

Least squares and motion. Nuno Vasconcelos ECE Department, UCSD Leas squares ad moo uo Vascocelos ECE Deparme UCSD Pla for oda oda we wll dscuss moo esmao hs s eresg wo was moo s ver useful as a cue for recogo segmeao compresso ec. s a grea eample of leas squares problem

More information

Cyclone. Anti-cyclone

Cyclone. Anti-cyclone Adveco Cycloe A-cycloe Lorez (963) Low dmesoal aracors. Uclear f hey are a good aalogy o he rue clmae sysem, bu hey have some appealg characerscs. Dscusso Is he al codo balaced? Is here a al adjusme

More information

For the plane motion of a rigid body, an additional equation is needed to specify the state of rotation of the body.

For the plane motion of a rigid body, an additional equation is needed to specify the state of rotation of the body. The kecs of rgd bodes reas he relaoshps bewee he exeral forces acg o a body ad he correspodg raslaoal ad roaoal moos of he body. he kecs of he parcle, we foud ha wo force equaos of moo were requred o defe

More information

Real-Time Systems. Example: scheduling using EDF. Feasibility analysis for EDF. Example: scheduling using EDF

Real-Time Systems. Example: scheduling using EDF. Feasibility analysis for EDF. Example: scheduling using EDF EDA/DIT6 Real-Tme Sysems, Chalmers/GU, 0/0 ecure # Updaed February, 0 Real-Tme Sysems Specfcao Problem: Assume a sysem wh asks accordg o he fgure below The mg properes of he asks are gve he able Ivesgae

More information

FORCED VIBRATION of MDOF SYSTEMS

FORCED VIBRATION of MDOF SYSTEMS FORCED VIBRAION of DOF SSES he respose of a N DOF sysem s govered by he marx equao of moo: ] u C] u K] u 1 h al codos u u0 ad u u 0. hs marx equao of moo represes a sysem of N smulaeous equaos u ad s me

More information

Available online Journal of Scientific and Engineering Research, 2014, 1(1): Research Article

Available online  Journal of Scientific and Engineering Research, 2014, 1(1): Research Article Avalable ole wwwjsaercom Joural o Scec ad Egeerg Research, 0, ():0-9 Research Arcle ISSN: 39-630 CODEN(USA): JSERBR NEW INFORMATION INEUALITIES ON DIFFERENCE OF GENERALIZED DIVERGENCES AND ITS APPLICATION

More information

Brownian Motion and Stochastic Calculus. Brownian Motion and Stochastic Calculus

Brownian Motion and Stochastic Calculus. Brownian Motion and Stochastic Calculus Browa Moo Sochasc Calculus Xogzh Che Uversy of Hawa a Maoa earme of Mahemacs Seember, 8 Absrac Ths oe s abou oob decomoso he bascs of Suare egrable margales Coes oob-meyer ecomoso Suare Iegrable Margales

More information

Partial Molar Properties of solutions

Partial Molar Properties of solutions Paral Molar Properes of soluos A soluo s a homogeeous mxure; ha s, a soluo s a oephase sysem wh more ha oe compoe. A homogeeous mxures of wo or more compoes he gas, lqud or sold phase The properes of a

More information

Final Exam Applied Econometrics

Final Exam Applied Econometrics Fal Eam Appled Ecoomercs. 0 Sppose we have he followg regresso resl: Depede Varable: SAT Sample: 437 Iclded observaos: 437 Whe heeroskedasc-cosse sadard errors & covarace Varable Coeffce Sd. Error -Sasc

More information

Optimal Eye Movement Strategies in Visual Search (Supplement)

Optimal Eye Movement Strategies in Visual Search (Supplement) Opmal Eye Moveme Sraeges Vsual Search (Suppleme) Jr Naemk ad Wlso S. Gesler Ceer for Percepual Sysems ad Deparme of Psychology, Uversy of exas a Aus, Aus X 787 Here we derve he deal searcher for he case

More information

Other Topics in Kernel Method Statistical Inference with Reproducing Kernel Hilbert Space

Other Topics in Kernel Method Statistical Inference with Reproducing Kernel Hilbert Space Oher Topcs Kerel Mehod Sascal Iferece wh Reproducg Kerel Hlber Space Kej Fukumzu Isue of Sascal Mahemacs, ROIS Deparme of Sascal Scece, Graduae Uversy for Advaced Sudes Sepember 6, 008 / Sascal Learg Theory

More information

FALL HOMEWORK NO. 6 - SOLUTION Problem 1.: Use the Storage-Indication Method to route the Input hydrograph tabulated below.

FALL HOMEWORK NO. 6 - SOLUTION Problem 1.: Use the Storage-Indication Method to route the Input hydrograph tabulated below. Jorge A. Ramírez HOMEWORK NO. 6 - SOLUTION Problem 1.: Use he Sorage-Idcao Mehod o roue he Ipu hydrograph abulaed below. Tme (h) Ipu Hydrograph (m 3 /s) Tme (h) Ipu Hydrograph (m 3 /s) 0 0 90 450 6 50

More information

Midterm Exam. Tuesday, September hour, 15 minutes

Midterm Exam. Tuesday, September hour, 15 minutes Ecoomcs of Growh, ECON560 Sa Fracsco Sae Uvers Mchael Bar Fall 203 Mderm Exam Tuesda, Sepember 24 hour, 5 mues Name: Isrucos. Ths s closed boo, closed oes exam. 2. No calculaors of a d are allowed. 3.

More information

International Journal Of Engineering And Computer Science ISSN: Volume 5 Issue 12 Dec. 2016, Page No.

International Journal Of Engineering And Computer Science ISSN: Volume 5 Issue 12 Dec. 2016, Page No. www.jecs. Ieraoal Joural Of Egeerg Ad Compuer Scece ISSN: 19-74 Volume 5 Issue 1 Dec. 16, Page No. 196-1974 Sofware Relably Model whe mulple errors occur a a me cludg a faul correco process K. Harshchadra

More information

Solution set Stat 471/Spring 06. Homework 2

Solution set Stat 471/Spring 06. Homework 2 oluo se a 47/prg 06 Homework a Whe he upper ragular elemes are suppressed due o smmer b Le Y Y Y Y A weep o he frs colum o oba: A ˆ b chagg he oao eg ad ec YY weep o he secod colum o oba: Aˆ YY weep o

More information

AML710 CAD LECTURE 12 CUBIC SPLINE CURVES. Cubic Splines Matrix formulation Normalised cubic splines Alternate end conditions Parabolic blending

AML710 CAD LECTURE 12 CUBIC SPLINE CURVES. Cubic Splines Matrix formulation Normalised cubic splines Alternate end conditions Parabolic blending CUIC SLINE CURVES Cubc Sples Marx formulao Normalsed cubc sples Alerae ed codos arabolc bledg AML7 CAD LECTURE CUIC SLINE The ame sple comes from he physcal srume sple drafsme use o produce curves A geeral

More information

EE 6885 Statistical Pattern Recognition

EE 6885 Statistical Pattern Recognition EE 6885 Sascal Paer Recogo Fall 005 Prof. Shh-Fu Chag hp://www.ee.columba.edu/~sfchag Lecure 5 (9//05 4- Readg Model Parameer Esmao ML Esmao, Chap. 3. Mure of Gaussa ad EM Referece Boo, HTF Chap. 8.5 Teboo,

More information

θ = θ Π Π Parametric counting process models θ θ θ Log-likelihood: Consider counting processes: Score functions:

θ = θ Π Π Parametric counting process models θ θ θ Log-likelihood: Consider counting processes: Score functions: Paramerc coug process models Cosder coug processes: N,,..., ha cou he occurreces of a eve of eres for dvduals Iesy processes: Lelhood λ ( ;,,..., N { } λ < Log-lelhood: l( log L( Score fucos: U ( l( log

More information

Complementary Tree Paired Domination in Graphs

Complementary Tree Paired Domination in Graphs IOSR Joural of Mahemacs (IOSR-JM) e-issn: 2278-5728, p-issn: 239-765X Volume 2, Issue 6 Ver II (Nov - Dec206), PP 26-3 wwwosrjouralsorg Complemeary Tree Pared Domao Graphs A Meeaksh, J Baskar Babujee 2

More information

The textbook expresses the stock price as the present discounted value of the dividend paid and the price of the stock next period.

The textbook expresses the stock price as the present discounted value of the dividend paid and the price of the stock next period. ublc Affars 974 Meze D. Ch Fall Socal Sceces 748 Uversy of Wscos-Madso Sock rces, News ad he Effce Markes Hypohess (rev d //) The rese Value Model Approach o Asse rcg The exbook expresses he sock prce

More information

Continuous Indexed Variable Systems

Continuous Indexed Variable Systems Ieraoal Joural o Compuaoal cece ad Mahemacs. IN 0974-389 Volume 3, Number 4 (20), pp. 40-409 Ieraoal Research Publcao House hp://www.rphouse.com Couous Idexed Varable ysems. Pouhassa ad F. Mohammad ghjeh

More information

Real-time Classification of Large Data Sets using Binary Knapsack

Real-time Classification of Large Data Sets using Binary Knapsack Real-me Classfcao of Large Daa Ses usg Bary Kapsack Reao Bru bru@ds.uroma. Uversy of Roma La Sapeza AIRO 004-35h ANNUAL CONFERENCE OF THE ITALIAN OPERATIONS RESEARCH Sepember 7-0, 004, Lecce, Ialy Oule

More information

The algebraic immunity of a class of correlation immune H Boolean functions

The algebraic immunity of a class of correlation immune H Boolean functions Ieraoal Coferece o Advaced Elecroc Scece ad Techology (AEST 06) The algebrac mmuy of a class of correlao mmue H Boolea fucos a Jgla Huag ad Zhuo Wag School of Elecrcal Egeerg Norhwes Uversy for Naoales

More information

Asymptotic Regional Boundary Observer in Distributed Parameter Systems via Sensors Structures

Asymptotic Regional Boundary Observer in Distributed Parameter Systems via Sensors Structures Sesors,, 37-5 sesors ISSN 44-8 by MDPI hp://www.mdp.e/sesors Asympoc Regoal Boudary Observer Dsrbued Parameer Sysems va Sesors Srucures Raheam Al-Saphory Sysems Theory Laboraory, Uversy of Perpga, 5, aveue

More information

Comparison of the Bayesian and Maximum Likelihood Estimation for Weibull Distribution

Comparison of the Bayesian and Maximum Likelihood Estimation for Weibull Distribution Joural of Mahemacs ad Sascs 6 (2): 1-14, 21 ISSN 1549-3644 21 Scece Publcaos Comarso of he Bayesa ad Maxmum Lkelhood Esmao for Webull Dsrbuo Al Omar Mohammed Ahmed, Hadeel Salm Al-Kuub ad Noor Akma Ibrahm

More information

Orbital Euclidean stability of the solutions of impulsive equations on the impulsive moments

Orbital Euclidean stability of the solutions of impulsive equations on the impulsive moments Pure ad Appled Mahemacs Joural 25 4(: -8 Publshed ole Jauary 23 25 (hp://wwwscecepublshggroupcom/j/pamj do: 648/jpamj254 ISSN: 2326-979 (Pr ISSN: 2326-982 (Ole Orbal ucldea sably of he soluos of mpulsve

More information

Pricing Asian Options with Fourier Convolution

Pricing Asian Options with Fourier Convolution Prcg Asa Opos wh Fourer Covoluo Cheg-Hsug Shu Deparme of Compuer Scece ad Iformao Egeerg Naoal Tawa Uversy Coes. Iroduco. Backgroud 3. The Fourer Covoluo Mehod 3. Seward ad Hodges facorzao 3. Re-ceerg

More information

On an algorithm of the dynamic reconstruction of inputs in systems with time-delay

On an algorithm of the dynamic reconstruction of inputs in systems with time-delay Ieraoal Joural of Advaces Appled Maemacs ad Mecacs Volume, Issue 2 : (23) pp. 53-64 Avalable ole a www.jaamm.com IJAAMM ISSN: 2347-2529 O a algorm of e dyamc recosruco of pus sysems w me-delay V. I. Maksmov

More information

Regression Approach to Parameter Estimation of an Exponential Software Reliability Model

Regression Approach to Parameter Estimation of an Exponential Software Reliability Model Amerca Joural of Theorecal ad Appled Sascs 06; 5(3): 80-86 hp://www.scecepublshggroup.com/j/ajas do: 0.648/j.ajas.060503. ISSN: 36-8999 (Pr); ISSN: 36-9006 (Ole) Regresso Approach o Parameer Esmao of a

More information

Mixed Integral Equation of Contact Problem in Position and Time

Mixed Integral Equation of Contact Problem in Position and Time Ieraoal Joural of Basc & Appled Sceces IJBAS-IJENS Vol: No: 3 ed Iegral Equao of Coac Problem Poso ad me. A. Abdou S. J. oaquel Deparme of ahemacs Faculy of Educao Aleadra Uversy Egyp Deparme of ahemacs

More information

COMPARISON OF ESTIMATORS OF PARAMETERS FOR THE RAYLEIGH DISTRIBUTION

COMPARISON OF ESTIMATORS OF PARAMETERS FOR THE RAYLEIGH DISTRIBUTION COMPARISON OF ESTIMATORS OF PARAMETERS FOR THE RAYLEIGH DISTRIBUTION Eldesoky E. Affy. Faculy of Eg. Shbee El kom Meoufa Uv. Key word : Raylegh dsrbuo, leas squares mehod, relave leas squares, leas absolue

More information

8. Queueing systems lect08.ppt S Introduction to Teletraffic Theory - Fall

8. Queueing systems lect08.ppt S Introduction to Teletraffic Theory - Fall 8. Queueg sysems lec8. S-38.45 - Iroduco o Teleraffc Theory - Fall 8. Queueg sysems Coes Refresher: Smle eleraffc model M/M/ server wag laces M/M/ servers wag laces 8. Queueg sysems Smle eleraffc model

More information

arxiv: v2 [cs.lg] 19 Dec 2016

arxiv: v2 [cs.lg] 19 Dec 2016 1 Sasfcg mul-armed bad problems Paul Reverdy, Vabhav Srvasava, ad Naom Ehrch Leoard arxv:1512.07638v2 [cs.lg] 19 Dec 2016 Absrac Sasfcg s a relaxao of maxmzg ad allows for less rsky decso makg he face

More information

The textbook expresses the stock price as the present discounted value of the dividend paid and the price of the stock next period.

The textbook expresses the stock price as the present discounted value of the dividend paid and the price of the stock next period. coomcs 435 Meze. Ch Fall 07 Socal Sceces 748 Uversy of Wscos-Madso Sock rces, News ad he ffce Markes Hypohess The rese Value Model Approach o Asse rcg The exbook expresses he sock prce as he prese dscoued

More information

Solution of Impulsive Differential Equations with Boundary Conditions in Terms of Integral Equations

Solution of Impulsive Differential Equations with Boundary Conditions in Terms of Integral Equations Joural of aheacs ad copuer Scece (4 39-38 Soluo of Ipulsve Dffereal Equaos wh Boudary Codos Ters of Iegral Equaos Arcle hsory: Receved Ocober 3 Acceped February 4 Avalable ole July 4 ohse Rabba Depare

More information

Interval Estimation. Consider a random variable X with a mean of X. Let X be distributed as X X

Interval Estimation. Consider a random variable X with a mean of X. Let X be distributed as X X ECON 37: Ecoomercs Hypohess Tesg Iervl Esmo Wh we hve doe so fr s o udersd how we c ob esmors of ecoomcs reloshp we wsh o sudy. The queso s how comforble re we wh our esmors? We frs exme how o produce

More information

The ray paths and travel times for multiple layers can be computed using ray-tracing, as demonstrated in Lab 3.

The ray paths and travel times for multiple layers can be computed using ray-tracing, as demonstrated in Lab 3. C. Trael me cures for mulple reflecors The ray pahs ad rael mes for mulple layers ca be compued usg ray-racg, as demosraed Lab. MATLAB scrp reflec_layers_.m performs smple ray racg. (m) ref(ms) ref(ms)

More information

Bianchi Type II Stiff Fluid Tilted Cosmological Model in General Relativity

Bianchi Type II Stiff Fluid Tilted Cosmological Model in General Relativity Ieraoal Joural of Mahemacs esearch. IN 0976-50 Volume 6, Number (0), pp. 6-7 Ieraoal esearch Publcao House hp://www.rphouse.com Bach ype II ff Flud led Cosmologcal Model Geeral elay B. L. Meea Deparme

More information

ENGINEERING solutions to decision-making problems are

ENGINEERING solutions to decision-making problems are 3788 IEEE TRANSACTIONS ON AUTOMATIC CONTROL, VOL. 62, NO. 8, AUGUST 2017 Sasfcg Mul-Armed Bad Problems Paul Reverdy, Member, IEEE, Vabhav Srvasava, ad Naom Ehrch Leoard, Fellow, IEEE Absrac Sasfcg s a

More information

Chapter 3: Maximum-Likelihood & Bayesian Parameter Estimation (part 1)

Chapter 3: Maximum-Likelihood & Bayesian Parameter Estimation (part 1) Aoucemes Reags o E-reserves Proec roosal ue oay Parameer Esmao Bomercs CSE 9-a Lecure 6 CSE9a Fall 6 CSE9a Fall 6 Paer Classfcao Chaer 3: Mamum-Lelhoo & Bayesa Parameer Esmao ar All maerals hese sles were

More information

1 Introduction and main results

1 Introduction and main results The Spacgs of Record Values Tefeg Jag ad Dag L 2 Ocober 0, 2006 Absrac. Le {R ; } be he sequece of record values geeraed from a sequece of..d. couous radom varables, wh dsrbuo from he expoeal famly. I

More information

Cyclically Interval Total Colorings of Cycles and Middle Graphs of Cycles

Cyclically Interval Total Colorings of Cycles and Middle Graphs of Cycles Ope Joural of Dsree Mahemas 2017 7 200-217 hp://wwwsrporg/joural/ojdm ISSN Ole: 2161-7643 ISSN Pr: 2161-7635 Cylally Ierval Toal Colorgs of Cyles Mddle Graphs of Cyles Yogqag Zhao 1 Shju Su 2 1 Shool of

More information

Density estimation III. Linear regression.

Density estimation III. Linear regression. Lecure 6 Mlos Hauskrec mlos@cs.p.eu 539 Seo Square Des esmao III. Lear regresso. Daa: Des esmao D { D D.. D} D a vecor of arbue values Obecve: r o esmae e uerlg rue probabl srbuo over varables X px usg

More information

Stability Criterion for BAM Neural Networks of Neutral- Type with Interval Time-Varying Delays

Stability Criterion for BAM Neural Networks of Neutral- Type with Interval Time-Varying Delays Avalable ole a www.scecedrec.com Proceda Egeerg 5 (0) 86 80 Advaced Corol Egeergad Iformao Scece Sably Crero for BAM Neural Neworks of Neural- ype wh Ierval me-varyg Delays Guoqua Lu a* Smo X. Yag ab a

More information

Density estimation III.

Density estimation III. Lecure 4 esy esmao III. Mlos Hauskrec mlos@cs..edu 539 Seo Square Oule Oule: esy esmao: Mamum lkelood ML Bayesa arameer esmaes MP Beroull dsrbuo. Bomal dsrbuo Mulomal dsrbuo Normal dsrbuo Eoeal famly Eoeal

More information

IMPROVED PORTFOLIO OPTIMIZATION MODEL WITH TRANSACTION COST AND MINIMAL TRANSACTION LOTS

IMPROVED PORTFOLIO OPTIMIZATION MODEL WITH TRANSACTION COST AND MINIMAL TRANSACTION LOTS Vol.7 No.4 (200) p73-78 Joural of Maageme Scece & Sascal Decso IMPROVED PORTFOLIO OPTIMIZATION MODEL WITH TRANSACTION COST AND MINIMAL TRANSACTION LOTS TIANXIANG YAO AND ZAIWU GONG College of Ecoomcs &

More information

Fully Fuzzy Linear Systems Solving Using MOLP

Fully Fuzzy Linear Systems Solving Using MOLP World Appled Sceces Joural 12 (12): 2268-2273, 2011 ISSN 1818-4952 IDOSI Publcaos, 2011 Fully Fuzzy Lear Sysems Solvg Usg MOLP Tofgh Allahvraloo ad Nasser Mkaelvad Deparme of Mahemacs, Islamc Azad Uversy,

More information

Extremal graph theory II: K t and K t,t

Extremal graph theory II: K t and K t,t Exremal graph heory II: K ad K, Lecure Graph Theory 06 EPFL Frak de Zeeuw I his lecure, we geeralize he wo mai heorems from he las lecure, from riagles K 3 o complee graphs K, ad from squares K, o complee

More information

Lecture 3 Topic 2: Distributions, hypothesis testing, and sample size determination

Lecture 3 Topic 2: Distributions, hypothesis testing, and sample size determination Lecure 3 Topc : Drbuo, hypohe eg, ad ample ze deermao The Sude - drbuo Coder a repeaed drawg of ample of ze from a ormal drbuo of mea. For each ample, compue,,, ad aoher ac,, where: The ac he devao of

More information

The Bernstein Operational Matrix of Integration

The Bernstein Operational Matrix of Integration Appled Mahemacal Sceces, Vol. 3, 29, o. 49, 2427-2436 he Berse Operaoal Marx of Iegrao Am K. Sgh, Vee K. Sgh, Om P. Sgh Deparme of Appled Mahemacs Isue of echology, Baaras Hdu Uversy Varaas -225, Ida Asrac

More information

Internet Appendix to: Idea Sharing and the Performance of Mutual Funds

Internet Appendix to: Idea Sharing and the Performance of Mutual Funds Coes Iere Appedx o: Idea harg ad he Perforace of Muual Fuds Jule Cujea IA. Proof of Lea A....................................... IA. Proof of Lea A.3...................................... IA.3 Proof of

More information

Density estimation. Density estimations. CS 2750 Machine Learning. Lecture 5. Milos Hauskrecht 5329 Sennott Square

Density estimation. Density estimations. CS 2750 Machine Learning. Lecture 5. Milos Hauskrecht 5329 Sennott Square Lecure 5 esy esmao Mlos Hauskrec mlos@cs..edu 539 Seo Square esy esmaos ocs: esy esmao: Mamum lkelood ML Bayesa arameer esmaes M Beroull dsrbuo. Bomal dsrbuo Mulomal dsrbuo Normal dsrbuo Eoeal famly Noaramerc

More information

NOTE ON SIMPLE AND LOGARITHMIC RETURN

NOTE ON SIMPLE AND LOGARITHMIC RETURN Appled udes Agrbusess ad Commerce AAC Ceer-r ublshg House, Debrece DOI:.94/AAC/27/-2/6 CIENIFIC AE NOE ON IME AND OGAIHMIC EUN aa Mskolcz Uversy of Debrece, Isue of Accoug ad Face mskolczpaa@gmal.com Absrac:

More information

ON TESTING EXPONENTIALITY AGAINST NBARFR LIFE DISTRIBUTIONS

ON TESTING EXPONENTIALITY AGAINST NBARFR LIFE DISTRIBUTIONS STATISTICA, ao LII,. 4, ON TESTING EPONENTIALITY AGAINST NBARR LIE DISTRIBUTIONS M. A. W. Mahmoud, N. A. Abdul Alm. INTRODUCTION AND DEINITIONS Tesg expoealy agas varous classes of lfe dsrbuos has go a

More information

Stabilization of LTI Switched Systems with Input Time Delay. Engineering Letters, 14:2, EL_14_2_14 (Advance online publication: 16 May 2007) Lin Lin

Stabilization of LTI Switched Systems with Input Time Delay. Engineering Letters, 14:2, EL_14_2_14 (Advance online publication: 16 May 2007) Lin Lin Egeerg Leers, 4:2, EL_4_2_4 (Advace ole publcao: 6 May 27) Sablzao of LTI Swched Sysems wh Ipu Tme Delay L L Absrac Ths paper deals wh sablzao of LTI swched sysems wh pu me delay. A descrpo of sysems sablzao

More information

General Complex Fuzzy Transformation Semigroups in Automata

General Complex Fuzzy Transformation Semigroups in Automata Joural of Advaces Compuer Research Quarerly pissn: 345-606x eissn: 345-6078 Sar Brach Islamc Azad Uversy Sar IRIra Vol 7 No May 06 Pages: 7-37 wwwacrausaracr Geeral Complex uzzy Trasformao Semgroups Auomaa

More information

The Signal, Variable System, and Transformation: A Personal Perspective

The Signal, Variable System, and Transformation: A Personal Perspective The Sgal Varable Syem ad Traformao: A Peroal Perpecve Sherv Erfa 35 Eex Hall Faculy of Egeerg Oule Of he Talk Iroduco Mahemacal Repreeao of yem Operaor Calculu Traformao Obervao O Laplace Traform SSB A

More information

Nonparametric Identification of the Production Functions

Nonparametric Identification of the Production Functions Proceedgs of he World Cogress o Egeerg 2011 Vol I, July 6-8, 2011, Lodo, U.. Noparamerc Idefcao of he Produco Fucos Geady oshk, ad Aa ayeva Absrac A class of sem-recursve kerel plug- esmaes of fucos depedg

More information

UNIVERSITY OF OSLO DEPARTMENT OF ECONOMICS

UNIVERSITY OF OSLO DEPARTMENT OF ECONOMICS UNIVERSITY OF OSLO DEPARTMENT OF ECONOMICS Postpoed exam: ECON430 Statstcs Date of exam: Jauary 0, 0 Tme for exam: 09:00 a.m. :00 oo The problem set covers 5 pages Resources allowed: All wrtte ad prted

More information

Quantitative Portfolio Theory & Performance Analysis

Quantitative Portfolio Theory & Performance Analysis 550.447 Quaave Porfolo heory & Performace Aalyss Week February 4 203 Coceps. Assgme For February 4 (hs Week) ead: A&L Chaper Iroduco & Chaper (PF Maageme Evrome) Chaper 2 ( Coceps) Seco (Basc eur Calculaos)

More information

Probability Bracket Notation and Probability Modeling. Xing M. Wang Sherman Visual Lab, Sunnyvale, CA 94087, USA. Abstract

Probability Bracket Notation and Probability Modeling. Xing M. Wang Sherman Visual Lab, Sunnyvale, CA 94087, USA. Abstract Probably Bracke Noao ad Probably Modelg Xg M. Wag Sherma Vsual Lab, Suyvale, CA 94087, USA Absrac Ispred by he Drac oao, a ew se of symbols, he Probably Bracke Noao (PBN) s proposed for probably modelg.

More information

Solving fuzzy linear programming problems with piecewise linear membership functions by the determination of a crisp maximizing decision

Solving fuzzy linear programming problems with piecewise linear membership functions by the determination of a crisp maximizing decision Frs Jo Cogress o Fuzzy ad Iellge Sysems Ferdows Uversy of Mashhad Ira 9-3 Aug 7 Iellge Sysems Scefc Socey of Ira Solvg fuzzy lear programmg problems wh pecewse lear membershp fucos by he deermao of a crsp

More information

Point Estimation: definition of estimators

Point Estimation: definition of estimators Pot Estmato: defto of estmators Pot estmator: ay fucto W (X,..., X ) of a data sample. The exercse of pot estmato s to use partcular fuctos of the data order to estmate certa ukow populato parameters.

More information

Rademacher Complexity. Examples

Rademacher Complexity. Examples Algorthmc Foudatos of Learg Lecture 3 Rademacher Complexty. Examples Lecturer: Patrck Rebesch Verso: October 16th 018 3.1 Itroducto I the last lecture we troduced the oto of Rademacher complexty ad showed

More information

( 1)u + r2i. f (x2i+1 ) +

( 1)u + r2i. f (x2i+1 ) + Malaya Joural of Maemak, Vol. 6, No., 6-76, 08 hps://do.org/0.667/mjm060/00 Geeral soluo ad geeralzed Ulam - Hyers sably of r ype dmesoal quadrac-cubc fucoal equao radom ormed spaces: Drec ad fxed po mehods

More information

On subsets of the hypercube with prescribed Hamming distances

On subsets of the hypercube with prescribed Hamming distances O subses of he hypercube wh prescrbed Hammg dsaces Hao Huag Oleksy Klurma Cosm Pohoaa Absrac A celebraed heorem of Klema exremal combaorcs saes ha a colleco of bary vecors {0, 1} wh dameer d has cardaly

More information

Random Generalized Bi-linear Mixed Variational-like Inequality for Random Fuzzy Mappings Hongxia Dai

Random Generalized Bi-linear Mixed Variational-like Inequality for Random Fuzzy Mappings Hongxia Dai Ro Geeralzed B-lear Mxed Varaoal-lke Iequaly for Ro Fuzzy Mappgs Hogxa Da Depare of Ecooc Maheacs Souhweser Uversy of Face Ecoocs Chegdu 674 P.R.Cha Absrac I h paper we roduce sudy a ew class of ro geeralzed

More information

RATIO ESTIMATORS USING CHARACTERISTICS OF POISSON DISTRIBUTION WITH APPLICATION TO EARTHQUAKE DATA

RATIO ESTIMATORS USING CHARACTERISTICS OF POISSON DISTRIBUTION WITH APPLICATION TO EARTHQUAKE DATA The 7 h Ieraoal as of Sascs ad Ecoomcs Prague Sepember 9-0 Absrac RATIO ESTIMATORS USING HARATERISTIS OF POISSON ISTRIBUTION WITH APPLIATION TO EARTHQUAKE ATA Gamze Özel Naural pulaos bolog geecs educao

More information

Fourth Order Runge-Kutta Method Based On Geometric Mean for Hybrid Fuzzy Initial Value Problems

Fourth Order Runge-Kutta Method Based On Geometric Mean for Hybrid Fuzzy Initial Value Problems IOSR Joural of Mahemacs (IOSR-JM) e-issn: 2278-5728, p-issn: 29-765X. Volume, Issue 2 Ver. II (Mar. - Apr. 27), PP 4-5 www.osrjourals.org Fourh Order Ruge-Kua Mehod Based O Geomerc Mea for Hybrd Fuzzy

More information

Efficient Estimators for Population Variance using Auxiliary Information

Efficient Estimators for Population Variance using Auxiliary Information Global Joural of Mahemacal cece: Theor ad Praccal. IN 97-3 Volume 3, Number (), pp. 39-37 Ieraoal Reearch Publcao Houe hp://www.rphoue.com Effce Emaor for Populao Varace ug Aular Iformao ubhah Kumar Yadav

More information

ClassificationofNonOscillatorySolutionsofNonlinearNeutralDelayImpulsiveDifferentialEquations

ClassificationofNonOscillatorySolutionsofNonlinearNeutralDelayImpulsiveDifferentialEquations Global Joural of Scece Froer Research: F Maheacs ad Decso Sceces Volue 8 Issue Verso. Year 8 Type: Double Bld Peer Revewed Ieraoal Research Joural Publsher: Global Jourals Ole ISSN: 49-466 & Pr ISSN: 975-5896

More information

STK4080/9080 Survival and event history analysis

STK4080/9080 Survival and event history analysis STK48/98 Survival ad eve hisory aalysis Marigales i discree ime Cosider a sochasic process The process M is a marigale if Lecure 3: Marigales ad oher sochasic processes i discree ime (recap) where (formally

More information

GENERALIZED METHOD OF LIE-ALGEBRAIC DISCRETE APPROXIMATIONS FOR SOLVING CAUCHY PROBLEMS WITH EVOLUTION EQUATION

GENERALIZED METHOD OF LIE-ALGEBRAIC DISCRETE APPROXIMATIONS FOR SOLVING CAUCHY PROBLEMS WITH EVOLUTION EQUATION Joural of Appled Maemacs ad ompuaoal Mecacs 24 3(2 5-62 GENERALIZED METHOD OF LIE-ALGEBRAI DISRETE APPROXIMATIONS FOR SOLVING AUHY PROBLEMS WITH EVOLUTION EQUATION Arkad Kdybaluk Iva Frako Naoal Uversy

More information

Photon Statistics. photons/s. Light beam stream of photons Difference with classical waves relies on different statistical properties

Photon Statistics. photons/s. Light beam stream of photons Difference with classical waves relies on different statistical properties hoo Sascs Lgh beam sream of phoos Dfferece wh classcal waves reles o dffere sascal properes Average cou rae correspods o Iesy of lgh beam Acual cou rae flucuaes from measureme o measureme The pulse aure

More information

An Efficient Dual to Ratio and Product Estimator of Population Variance in Sample Surveys

An Efficient Dual to Ratio and Product Estimator of Population Variance in Sample Surveys "cece as True Here" Joural of Mahemacs ad ascal cece, Volume 06, 78-88 cece gpos Publshg A Effce Dual o Rao ad Produc Esmaor of Populao Varace ample urves ubhash Kumar Yadav Deparme of Mahemacs ad ascs

More information