Asymptotic Efficiencies of the MLE Based on Bivariate Record Values from Bivariate Normal Distribution
|
|
- Sharleen Dickerson
- 5 years ago
- Views:
Transcription
1 JIRSS 013 Vol. 1, No., pp 35-5 Downloaded from jrss.rstat.r at 1: on Monday September 17th 018 Asymptotc ffcences of the ML Based on Bvarate Record Values from Bvarate Normal Dstrbuton Morteza Amn 1, Jafar Ahmad 1 Department of Statstcs, School of Mathematcs, Statstcs and Computer Scence, College of Scence, Unversty of Tehran, Iran. Department of Statstcs, Ferdows Unversty of Mashhad, Iran. Abstract. Maxmum lkelhood ML estmaton based on bvarate record data s consdered as the general nference problem. Assume that the process of observng k records s repeated m tmes, ndependently. The asymptotc propertes ncludng consstency and asymptotc normalty of the Maxmum Lkelhood ML estmates of parameters of the underlyng dstrbuton s then establshed, when m s large enough. The bvarate normal dstrbuton s consdered as an hghly applcable example n order to estmate the parameter θ = µ 1,, µ, σ by ML method of estmaton based on mk bvarate record data. Asymptotc varances of the ML estmators are calculated by dervng the Fsher nformaton matrx about θ contaned n the vector of the frst k bvarate record data. As another applcaton, we concerned the problem of breakng boards of Glck 1978, Amer. Math. Monthly, 85, -6 by consderng three dfferent samplng schemes of breakng boards and we computed the relatve asymptotc effcences of ML estmators based on these three types of data. Keywords. Addtvty, bvarate dstrbuton, Fsher nformaton matrx, nverse samplng, Jensen s nequalty. MSC: 6F1, 6N0, 6F10. Morteza Amn morteza.amn@ut.ac.r, Jafar Ahmadahmad-j@um.ac.r Receved: Aprl 01; Accepted: September 01
2 36 Amn and Ahmad 1 Introducton Downloaded from jrss.rstat.r at 1: on Monday September 17th 018 In a sequence of par-wse random varables, the values of one varable that exceed or fall below the prevous values of that varable are called record values and the correspondng value of other varable to each exceedng value s the concomtant of that record value. The bvarate sequence s then termed the sequence of bvarate record values. The reader s referred to Arnold et al for more detals on unvarate and bvarate record data. stmaton of parameters of bvarate normal dstrbuton based on bvarate ordered random varables s concerned by a few authors. Harrell and Sen 1979, Qnyng He and Nagaraja 009 and the references theren, consder the problem of estmaton based on bvarate order statstcs under the bvarate normal dstrbuton. Chacko and Thomas 008 obtaned best lnear unbased estmators of the parameters of the bvarate normal dstrbuton n terms of only concomtants of record values. The problem of dervng Fsher nformaton contaned n bvarate record values was studed by Amn and Ahmad 007, 008. Amn and Ahmad 009 studed the addtonal nformaton by consderng nter-record tmes. In ths paper we consder ML estmaton based on bvarate record data. Snce for a sngle record-breakng sample, the number of record data are rare, t s napplcable to nvestgatng the asymptotc propertes of the estmators when the number of records n a sequence gets large. In nonparametrc settng, Samanego and Whtaker 1988 extend the sngle sample results to the mult-sample case and studed the consstency of the nonparametrc maxmum-lkelhood estmator of the samplng dstrbuton. In ths samplng scheme, one repeat an nverse samplng scheme for achevng record data m tmes. The asymptotc propertes, then, can be concerned by lettng m to get large. A balanced scheme ncludes m ndependent sequences each wth the frst k record data. We show that n such a scheme, the ML estmators are consstent and asymptotcally normal as m tends to nfnty. One of the most mportant bvarate dstrbutons n statstcal nference s the bvarate normal dstrbuton. As an applcaton of the proposed estmaton we estmate the parameter θ = µ 1,, µ, σ by ML estmaton method based on mk bvarate record data. Asymptotc varances of the ML estmators are calculated by dervng the Fsher nformaton matrx about θ contaned n the vector of the frst k bvarate record data.
3 Asymptotc ffcences of the ML Based on Downloaded from jrss.rstat.r at 1: on Monday September 17th 018 As another applcaton of the aforementoned nferental problem consder lfe testng problems n whch testng of an tem s destructve and costly. If the tems are expensve, one can set up the experment so that only the low lfe unts are destroyed. As an example, one may consder the example of testng the breakng strength of wooden beams cted n Gulat and Padgett 1994 a, b and Glck 1978, n whch beams are replaced when they do not break untl the pressure reaches the mnmum prevously observed breakng tme. In other words, only the lower record values are observed. Now, suppose that we want to ft a model for the relatonshp between the breakng strength as the response varable and some other characterstcs of wooden beams as predctors such as elastcty. Here our only observatons are the lower records of breakng strength and ther assocated concomtants. In such stuatons the problem of estmaton based on record values and ther concomtants s n a hgh degree of mportance. Suppose we want to observe the strength X and elastcty Y of mk beams by a destructve lfe test. One can consder, the followng three dfferent samplng schemes for testng the wooden beams: A: Test the beams by replacng them when they do not break untl the pressure reaches the mnmum prevously observed breakng tme. Contnue the experment untl breakng k beams. Repeat the whole experment m tmes. B: Perform test A and also save the number of beams remanng unbroken between the th and the + 1 th breakng beams n each replcaton. C: Smply test mk beams untl breakng. Test A and B are smlar tests, except that we observe an addtonal varable whch are record tmes. Indeed, n test B, we consder the addtonal nformaton that the strength of the beams remanng unbroken between the th and the + 1 th breakng beams s greater than the strength of the th breakng beam. In test C the expermenter observe a bvarate random sample of sze mk. Although, n tests A and B, more beams are tested, the number of destroyed beams n the three tests s equal. However, the tests A and B are more economcal tests, snce the more low lfe tems are destroyed. The queston s that whch samplng scheme leads to more precse estmators of populaton parameters?. Here, the relatve asymptotc effcences of the ML estmators, based on the above three types of
4 38 Amn and Ahmad Downloaded from jrss.rstat.r at 1: on Monday September 17th 018 data, can be calculated usng the Fsher nformaton matrx about θ contaned n the vector of the frst k bvarate record data. The paper s organzed as follows. Secton contans some notatons. In Secton 3, the man results of ML estmaton and lmt theorems are proposed. Secton 4, concerns the ML estmaton of the parameters of the bvarate normal dstrbuton, usng bvarate record values. Fnally, we calculate the Fsher nformaton matrces and asymptotc relatve effcences. Notatons Let {X, Y, 1} be a sequence of..d. par-wse random varables wth the absolutely contnuous cumulatve dstrbuton functon cdf F X,Y x, y; θ and the correspondng pdf f X,Y x, y; θ. Also f X x; θ and F X x; θ denote the margnal pdf and cdf of X, respectvely and F X x; θ = 1 F X x; θ. The sequence of lower records and ther concomtants s defned as R, R ] = X T, Y T, 1, where T 1 = 1 wth probablty one and for, T = mn{j : j > T 1, X j < X T 1 }. Suppose = T +1 T 1, = 1,,..., k 1, k = 0, are the number of trals needed to obtan new records, whch are called nter-record tmes. Let us denote and Rk = R 1,..., R k, k = 1,..., k, Ck = R 1],..., R k], R m = {R 1j,..., R kj, j = 1,..., m}, C m = {R 1]j,..., R k]j, j = 1,..., m}, m = { 1j,..., kj, j = 1,..., m}, where R j, R ]j, j stands for the th record data n the jth sequence, = 1,..., k, j = 1,..., m. The jont pdf of R m and C m s see Arnold et al., 1998 f Rm,C m r, c; θ = m k j=1 =1 k 1 f X,Y r j, c j ; θ/ F X r j ; θ]. 1 =1
5 Asymptotc ffcences of the ML Based on Usng 1 the jont pdf of R m, C m and m s gven by m k f Rm,C m, m r, c, δ; θ = f X,Y r j, c j ; θ{ F X r j ; θ} δ j. j=1 =1 Downloaded from jrss.rstat.r at 1: on Monday September 17th 018 So, the condtonal probablty mass functon of m gven R m, C m s gven by f m R m,c m δ r, c; θ = m k 1 F X r j ; θ] δ j F X r j ; θ. 3 j=1 =1 We consder lower bvarate records to derve the results of ths paper. Smlar results also hold for the case of upper bvarate record data. Hereafter, we wll call lower records, smply, records. 3 Man Results Suppose that the assumptons of Secton hold. Also, let θ = θ 1,..., θ l Θ R l and the data nvolves R m, C m, m. Usng 1 and, the lkelhood equatons based on bvarate records only are j=1 =1 θ s log f X,Y r j, c j ; θ k 1 j=1 =1 θ s log F r j ; θ = 0, s = 1,..., l. Also the lkelhood equatons based on bvarate records and nter-record tmes are gven by j=1 =1 log f X,Y r j, c j ; θ+ θ s j=1 =1 δ j log θ F r j ; θ = 0, s = 1,..., l. s The followng theorems establsh consstency and asymptotc normalty of the roots of the above equatons f exst, as m tends to nfnty. Theorem 3.1. Let ˆθ m be the ML of θ based on m ndependent sequences of the frst k bvarate record values, then ˆθ m s a consstent estmator for θ as m. Proof. Let θ 0 be the value of θ. Snce ˆθ m s the ML estmator of θ, we have log f Rm,C m ϱ m, ς m ; ˆθ m log f Rm,C m ϱ m, ς m ; θ, for all θ Θ. 4
6 40 Amn and Ahmad On the other hand by Jensen s nequalty we have θ0 log f R,Cr, s; θ fr,c r, s; θ log θ0 f R,C r, s; θ 0 f R,C r, s; θ 0 = 0. Downloaded from jrss.rstat.r at 1: on Monday September 17th 018 So θ0 log f R,C r, s; θ θ0 log f R,C r, s; θ 0. Snce for all θ Θ 1 m log f R m,c m ϱ m, ς m ; θ tends almost everywhere to θ0 log f R,C r, s; θ as m, hence wth probablty one and for large enough m, we have 1 m log f R m,c m ϱ m, ς m ; θ 1 m log f R m,c m ϱ m, ς m ; θ 0, for all θ Θ. 5 Settng θ = θ 0 n 4 and θ = ˆθ m n 5, t s deduced that ˆθ m tends to θ 0 wth probablty one, as m. Theorem 3.. Let ˆθ m be as n Theorem 3.1, then the asymptotc dstrbuton of mˆθ m θ s N l 0, I 1 Rk,Ck θ as m, where θ s the nverse of Fsher nformaton matrx n Rk, Ck. I 1 Rk,Ck Proof. The Taylor s expanson of θ r log f Rm,Cm ϱ m, ς m ; θ for r = θ=ˆθ 1,..., l around an arbtrarly θ 1 s equal to log f Rm,C θ m ϱ m, ς m ; θ r θ=ˆθ = log f Rm,C θ m ϱ m, ς m ; θ r θ=θ1 l + ˆθ s θ 1 log f Rm,C θ r θ m ϱ m, ς m ; θ, s s=1 θ=θ where θ s a sequence n Θ whch tends n probablty to θ 1. The left hand sde of the above equalty s equal to zero and hence θ log f R m,c m ϱ m, ς m ; θ θ=θ1 l = ˆθ s θ 1 log f Rm,C θθ m ϱ m, ς m ; θ s s=1 = mˆθ θ 1 I Rk,Ck θ 1. θ=θ
7 Asymptotc ffcences of the ML Based on Downloaded from jrss.rstat.r at 1: on Monday September 17th 018 On the other hand, θ log f R m,c m ϱ m, ς m ; θ θ=θ1 has a mean equal to zero and a varance equal to mi Rk,Ck θ 1. So, by strong low of large numbers, we have 1 m θ log f R m,c m ϱ m, ς m ; θ = m 1 m θ=θ1 θ log f R m,c m ϱ m, ς m ; θ 0 θ=θ1 has an asymptotc dstrbuton N l 0, I Rk,Ck θ 1 as m. Hence, mˆθm θ = I 1 Rk,Ck θ 1 m θ log f R m,c m ϱ m, ς m ; θ has an asymptotc dstrbuton N l 0, I 1 Rk,Ck θ as m. Remark 3.1. Smlar asymptotc results as presented n Theorems 3.1 and 3., are hold for ML estmators based on bvarate records and ther nter-record tmes by replacng f Rm,C m ϱ m, ς m ; θ wth f Rm,C m, m ϱ m, ς m, δ m ; θ and I R,C θ wth I R,C, θ n Theorems 3.1 and Bvarate Normal Parameter stmaton Suppose that m ndependent sequences each wth k bvarate records are observed and the samplng dstrbuton s bvarate normal dstrbuton wth parameter θ = µ 1,, µ, σ, ρ. Let R j, R ]j denote the th bvarate record n the jth sequence. Also, denote the standard hazard rate functon and standard reverse hazard rate functon of X, respectvely, by h 0 x = f X x; θ 0 / F X x, θ 0 and r 0 x = f X x; θ 0 /F X x, θ 0, n whch θ 0 = 0, 1,, µ, σ, ρ. Our am s to obtan the ML of θ based on bvarate records. The lkelhood equatons for bvarate record values and record tmes are as follows: j=1 =1 Rj µ 1 ρ m j=1 =1 1 ρ R]j µ j=1 =1 σ Rj µ 1 h 0 = 0, 6
8 Downloaded from jrss.rstat.r at 1: on Monday September 17th Amn and Ahmad R ]j mkµ ρσ R σ j mkµ 1 = 0, 1 j=1 =1 j=1 =1 m k j=1 =1 ρ mk R j µ 1 m k j=1 =1 R ]j µ σ 1 σ +C θρ + 1 mkρ 3 = 0, where C θ = m j=1 k =1 Rj µ 1 R]j µ /σ1 σ, and +1 ρ mk1 ρ j=1 =1 mk1 ρ j=1 =1 R j µ 1 h 0 j=1 =1 Rj µ 1 + ρc θ Rj µ 1 = 0, 7 R]j µ + ρc θ = 0. For the case of bvarate record values wthout record tmes, the above equatons reman true except that 6 and 7 are replaced, respectvely, wth Rj µ 1 ρ m R]j µ σ and j=1 =1 1 ρ mk1 ρ +1 ρ j=1 =1 j=1 =1 j=1 =1 σ j=1 =1 r 0 Rj µ 1 = 0, Rj µ 1 + ρc θ R j µ 1 r 0 Rj µ 1 = 0. The roots of the above lkelhood equatons have no closed form and these equatons have to be solved numercally. However, t s not dffcult to carry out an numercal computaton usng mathematcal packages, e.g. Maple or Mathematca.
9 Asymptotc ffcences of the ML Based on Fsher Informaton Matrx Downloaded from jrss.rstat.r at 1: on Monday September 17th 018 In order to compute the asymptotc relatve effcences of the ML estmator of θ based on m ndependent sequences each wth k bvarate records and tmes denoted wth RCT, wth correspondng estmator based on smlar mk bvarate records wthout tmes denoted wth RC, and the estmator based on a bvarate random sample of sze mk denoted wth IID, we derve the Fsher nformaton matrces n these three data bases. We have ki X,Y θ = I j ;, j = 1..., 5, such that I j = I j ; j and I 11 = k σ 1 1 ρ, I 1 = 0, I 13 = I = k1 ρ / σ ρ, I 3 = 0, I 4 = and I 33 = kρ σ 1 ρ, I 14 = 0, I 15 = 0, kρ 4σ 1 σ 1 ρ, I 5 = k σ 1 ρ, I 34 = 0, I 35 = 0, I 44 = k1 ρ / σ 4 1 ρ, I 45 = I 55 = k1 + ρ 1 ρ. kρ σ 1 ρ, kρ σ 1 1 ρ, We denote some moments of bvarate records from standard bvarate normal dstrbuton as follows α ] = R 0,1 R 0,1 ], α ] = R 0,1 ] and α = R, where R 0,1, R 0,1 ] s the th bvarate record from standard normal dstrbuton. Suppose the dstrbuton of X, Y s standard bvarate normal wth correlaton ρ for =1,,..., then Y = ρx + ϵ, 8 where the X and the ϵ are mutually ndependent and ϵ are normal dstrbuted wth zero mean and varance equal to 1 ρ. So by consderng X-record sequence we have for 1 R 0,1 ] = ρr 0,1 + ϵ ], 9
10 44 Amn and Ahmad where ϵ ] denotes the partcular ϵ assocated wth R 0,1. Snce X and ϵ are ndependent so the sequence R 0,1 s ndependent of ϵ ], the later beng mutually ndependent, each wth the same dstrbuton as ϵ. So we can conclude from 9 that for any fnte functon g. Downloaded from jrss.rstat.r at 1: on Monday September 17th 018 R 0,1 ] gr0,1 = ρr 0,1 gr 0,1 and R 0,1 ] gr 0,1 = ρ R 0,1 gr 0,1 + 1 ρ gr 0,1. specally we have R 0,1 ] R0,1 = α ] = ρα and R 0,1 ] ] = α ] = ρ α + 1 ρ. The log-lkelhood functon of the frst k bvarate records s equal to where lθ; Rk, Ck = L θ; R, R ] log Φ =1 =1 R µ 1 L θ; R, R ] = 1 logσ1 1 ρ 1 ρ R µ 1 R] µ + σ σ ] R µ 1 R] µ ρ., Hence, we obtan I Rk,Ck θ = I j ;, j = 1..., 5, such that I j = I j ; j and I 11 = =1 µ L θ; R, R ] 1 1 σ1 r 0R 0,1 = 1 k 1 k σ1 1 ρ + =1 R 0,1 =1 r 0 R 0,1 r0r 0,1 =1 ],
11 Downloaded from jrss.rstat.r at 1: on Monday September 17th 018 Asymptotc ffcences of the ML Based on and I = =1 1 = 1 σ 4 1 4σ = 1 σ σ1 L θ; R, R ] 3 =1 k ρ =1 R 0,1 r 0 R 0,1 { + =1 α =1 R 0,1 r 0 R 0,1 1 4 k + 1 3/4ρ 1 ρ =1 =1 α R 0,1 r 0R 0,1 } 3 4 ρ α ] =1 R 0,1 r 0 R 0, =1 ] 1 R 0,1 r 4 0R 0,1, =1 I 33 = µ L θ; R, R ] = =1 I 44 = = 1 σ 4 = 1 σ 4 =1 σ L θ; R, R ] { k ρ =1 k + ρ 41 ρ =1 α α ] ]. =1 ] R 0,1 r 0R 0,1 R 0,1 3 r 0 R 0,1 k σ 1 ρ, }] 3 4 ρ α ] =1 ] Amn and Ahmad 007 showed that I 55 = 1 ρ k =1 α + kρ 1 ρ.
12 46 Amn and Ahmad Downloaded from jrss.rstat.r at 1: on Monday September 17th 018 Furthermore, we have I 1 = =1 1 σ 3 1 µ 1 σ1 L θ; R, R ] k 1 r 0 R 0,1 + R 0,1 r 0R 0,1 =1 { = 1 1 σ1 3 1 ρ α ρ { k 1 =1 =1 } α ] =1 1 r 0 R 0,1 + =1 =1 { = 1 1 ρ / σ1 3 1 ρ α 1 I 13 = I 14 = I 15 = I 3 = I 4 = I 5 = I 34 = I 35 = =1 =1 R 0,1 r 0 R 0,1 + =1 =1 =1 =1 =1 =1 =1 =1 L = µ 1 µ µ 1 σ L = µ 1 ρ L = σ1 µ L = σ1 L = σ σ1 = ρl µ σ L = µ ρ L = ] R 0,1 r 0R 0,1 =1 =1 r 0 R 0,1 }] R 0,1 r 0R 0,1 kρ σ 1 ρ, ρ σ 1 ρ ρ 1 ρ α, =1 α, =1 ρ σ 1 σ 1 ρ ρ 4σ 1 σ 1 ρ ρ σ 1 1 ρ ρ σ 3 1 ρ 1 σ 1 ρ =1 α, =1 =1 α, α, =1 α, =1 α, }],
13 Asymptotc ffcences of the ML Based on and Downloaded from jrss.rstat.r at 1: on Monday September 17th 018 I 45 = =1 σ = ρl ] ρ σ 1 α ρ k. =1 The Fsher nformaton matrx n the frst k bvarate records and record tmes s equal to I Rk,Ck, k θ = I j ;, j = 1..., 5, such that I j = I j ; j. Snce the condtonal pdf f k Rk,Ck depends only on µ 1 and and s free of other parameters, we have for {, j; 1 5, 1 j 5, j} {1, 1,,, 1, }, Furthermore I j = I j θ θj log f R,C = I j. I 11 = 1 k σ1 1 ρ =1 = 1 k σ1 1 ρ + =1 { I = 1 k σ ρ =1 3 R 0,1 1 ΦR 0,1 4 =1 ΦR 0,1 1 R 0,1 1 ΦR0,1 4 =1 k + 1 3/4ρ 1 ρ = 1 σ =1 1 ΦR 0,1 ΦR 0,1 R 0,1 r 0 R 0,1 ΦR 0,1 =1 R 0,1 3 r 0 R 0,1 α h 0 R 0,1 h 0R 0,1 + =1 ] } 3 ρ α ] h 0R 0,1 α =1 =1 =1 ] ] r 0 R 0,1 h 0 R 0,1, R 0,1 r 0 R 0,1 ] R 0,1 r 0 R 0,1 h 0 R 0,1,
14 48 Amn and Ahmad Downloaded from jrss.rstat.r at 1: on Monday September 17th 018 and I 1 = I 1 µ 1 σ1 log f R,C δ r, s = I 1 R µ 1 log Φ =1 = 1 σ 3 1 =1 =1 1 ΦR 0,1 1 ρ / 1 ρ ΦR 0,1 µ 1 σ1 R µ 1 µ 1 σ1 log 1 Φ α 1 =1 { k 1 =1 =1 r 0 R 0,1 R 0,1 r 0 R 0,1 R 0,1 h 0 R 0,1 r 0 R 0,1 4. Asymptotc relatve effcences In ths secton, we try to fnd an answer for the queston whch samplng scheme of A, B or C, ntroduced n Secton 1 leads to more precse estmators of populaton parameters?, for the specal case of bvarate normal dstrbuton. We use the asymptotc varances of the ML estmators whch are the trace tems of the nverse of the Fsher nformaton matrces contaned n the three types of data. Snce µ 1 and µ are locaton parameters, the Fsher nformaton matrces are free of µ 1 and µ. It can be seen that I 1 1, 1 s multple of σ1, I 1, s multple of σ1 4, I 1 3, 3 s multple of σ, I 1 4, 4 s multple of σ 4 and I 1 5, 5 s free of and σ, for I be each of the three mentoned nformaton matrces, where I 1, s the element of the th row and the th column of the matrx I 1. Thus, the asymptotc relatve effcency denoted by AR gven below does not depend on µ and σ, = 1,. Hence, wthout loss of generalty, we take = σ = 1 n evaluatng I 1 Rk,Ck }] θ, I 1 Rk,Ck, k θ and ki X,Y 1 θ usng the above formulas. For smplcty, we denote the ML estmate of θ based on Rk, Ck by ˆθ A, the ML of θ based on Rk, Ck, k by ˆθ B and the ML of θ based on the bvarate random sample of sze mk by ˆθ C. Asymptotc effcency of the ML usng bvarate record values and tmes wth respect to w.r.t. that usng bvarate random sample of the same sze, and AR of the ML usng bvarate record values and tmes w.r.t. that usng bvarate record.
15 Asymptotc ffcences of the ML Based on values wthout tmes are as follows: ARˆθ,B ; ˆθ,C = ki X,Y 1,, ; I 1 Rk,Ck, k Downloaded from jrss.rstat.r at 1: on Monday September 17th 018 ARˆθ,B ; ˆθ,A = I 1 Rk,Ck, I 1 for = 1,..., 5. Rk,Ck, k,, Tables 1 to 5 show the values of AR of µ 1,, µ, σ and ρ, respectvely. These are smlar for negatve and postve values of ρ. Snce the values varatons for Tables 1 and w.r.t. ρ are very neglgble, they are shown n 8 decmal places. The other tables values are shown n 4 decmal places. Table 1: The values of ARˆµ 1,B ; ˆµ 1,C ARˆµ 1,B ; ˆµ 1,A. ρ k Table : The values of ARˆ,B ; ˆ,C ARˆ,B ; ˆ,A. ρ k
16 50 Amn and Ahmad Table 3: The values of ARˆµ,B ; ˆµ,C ARˆµ,B ; ˆµ,A. Downloaded from jrss.rstat.r at 1: on Monday September 17th 018 ρ k Table 4: The values of ARˆσ,B ; ˆσ,C ARˆσ,B ; ˆσ,A. ρ k Table 5: The values of ARˆρ B ; ˆρ C ARˆρ B ; ˆρ A. ρ k
17 Asymptotc ffcences of the ML Based on Downloaded from jrss.rstat.r at 1: on Monday September 17th 018 As we can see from the Tables 1 to 5, consderng nter-record tmes along by bvarate records, provdes a dstnct mprove n estmaton of θ. Also, plans A and B lead to more precse estmators of scale and correlaton parameters of bvarate normal dstrbuton. On the other hand, locaton parameters estmates are more precse based on plan C. Furthermore, smaller values of k, for locaton parameters, and greater values of k, for scale and correlaton parameters, lead to mprovement of the estmators based on plans A and B relatve to the estmators based on plan C. Also, ARˆσ,B ; ˆσ,C s relatve to ρ and ARˆµ,B ; ˆµ,C and ARˆρ B ; ˆρ C have reverse relaton wth ρ. References Abo-leneen, Z. A. and Nagaraja, H. N. 00, Fsher nformaton n an order statstc and ts concomtant. Ann. Inst. Statst. Math., 543, Abo-leneen, Z. A. 007, Fsher nformaton n generalzed order statstcs and ther concomtants. J. Stat. Theory Appl., 6, Abo-leneen, Z. A. and Nagaraja, H. N. 008, Fsher nformaton n order statstcs and ther concomtants n bvarate censored samples. Metrka, 67, Al-Saleh, M. F. and Al-Ananbeh, A. M. 007, stmaton of the means of the bvarate normal usng movng extreme ranked set samplng wth concomtant varable. Stat. Papers, 48, Amn, M. and Ahmad, J. 007, Fsher nformaton n record values and ther concomtants about the dependence and correlaton parameters. Statst. Probab. Lett., 77, Amn, M. and Ahmad, J. 008, Comparng Fsher nformaton n record values and ther concomtants wth random observatons. Statstcs, 4, Arnold, B. C., Balakrshnan, N., and Nagaraja, H. N., 1998, Records. New York: John Wley & Sons.
18 5 Amn and Ahmad Chacko, M., and Thomas P. Y. 008, stmaton of parameters of bvarate normal dstrbuton usng concomtants of record values. Stat. Papers, 49, Downloaded from jrss.rstat.r at 1: on Monday September 17th 018 Glck, N. 1978, Breakng record and breakng boards. Amer. Math. Monthly, 85, -6. Gulat, S. and Padgett, W. J. 1994a, Smooth nonparametrc estmaton of the dstrbuton and densty functons from record-breakng data. Commun. Stat. Theory. Meth., 3, Gulat, S. and Padgett, W. J. 1994b, Nonparametrc quantle estmaton from record-breakng data. Austral. Statst., 36, Harrell F.. and Sen P. K. 1979, Statstcal nference for censored bvarate normal dstrbutons based on nduced order statstcs. Bometrka, 66, He, Q. and Nagaraja, H. N. 009, Correlaton stmaton Usng Concomtants of Order Statstcs from Bvarate Normal Samples. Commun. Statst. Theory Meth., 381, Samanego, F. J. and Whtaker, L. R. 1988, On estmatng populaton characterstcs from record-breakng observatons, II: Nonparametrc results. Nav. Res. Logst. Q., 35, 1-36.
3.1 Expectation of Functions of Several Random Variables. )' be a k-dimensional discrete or continuous random vector, with joint PMF p (, E X E X1 E X
Statstcs 1: Probablty Theory II 37 3 EPECTATION OF SEVERAL RANDOM VARIABLES As n Probablty Theory I, the nterest n most stuatons les not on the actual dstrbuton of a random vector, but rather on a number
More informationComputing MLE Bias Empirically
Computng MLE Bas Emprcally Kar Wa Lm Australan atonal Unversty January 3, 27 Abstract Ths note studes the bas arses from the MLE estmate of the rate parameter and the mean parameter of an exponental dstrbuton.
More informationEcon107 Applied Econometrics Topic 3: Classical Model (Studenmund, Chapter 4)
I. Classcal Assumptons Econ7 Appled Econometrcs Topc 3: Classcal Model (Studenmund, Chapter 4) We have defned OLS and studed some algebrac propertes of OLS. In ths topc we wll study statstcal propertes
More informationEstimation: Part 2. Chapter GREG estimation
Chapter 9 Estmaton: Part 2 9. GREG estmaton In Chapter 8, we have seen that the regresson estmator s an effcent estmator when there s a lnear relatonshp between y and x. In ths chapter, we generalzed the
More informationFirst Year Examination Department of Statistics, University of Florida
Frst Year Examnaton Department of Statstcs, Unversty of Florda May 7, 010, 8:00 am - 1:00 noon Instructons: 1. You have four hours to answer questons n ths examnaton.. You must show your work to receve
More informationThe Multiple Classical Linear Regression Model (CLRM): Specification and Assumptions. 1. Introduction
ECONOMICS 5* -- NOTE (Summary) ECON 5* -- NOTE The Multple Classcal Lnear Regresson Model (CLRM): Specfcaton and Assumptons. Introducton CLRM stands for the Classcal Lnear Regresson Model. The CLRM s also
More informationParametric fractional imputation for missing data analysis. Jae Kwang Kim Survey Working Group Seminar March 29, 2010
Parametrc fractonal mputaton for mssng data analyss Jae Kwang Km Survey Workng Group Semnar March 29, 2010 1 Outlne Introducton Proposed method Fractonal mputaton Approxmaton Varance estmaton Multple mputaton
More informationComposite Hypotheses testing
Composte ypotheses testng In many hypothess testng problems there are many possble dstrbutons that can occur under each of the hypotheses. The output of the source s a set of parameters (ponts n a parameter
More informationDurban Watson for Testing the Lack-of-Fit of Polynomial Regression Models without Replications
Durban Watson for Testng the Lack-of-Ft of Polynomal Regresson Models wthout Replcatons Ruba A. Alyaf, Maha A. Omar, Abdullah A. Al-Shha ralyaf@ksu.edu.sa, maomar@ksu.edu.sa, aalshha@ksu.edu.sa Department
More informationLINEAR REGRESSION ANALYSIS. MODULE IX Lecture Multicollinearity
LINEAR REGRESSION ANALYSIS MODULE IX Lecture - 30 Multcollnearty Dr. Shalabh Department of Mathematcs and Statstcs Indan Insttute of Technology Kanpur 2 Remedes for multcollnearty Varous technques have
More informationA Robust Method for Calculating the Correlation Coefficient
A Robust Method for Calculatng the Correlaton Coeffcent E.B. Nven and C. V. Deutsch Relatonshps between prmary and secondary data are frequently quantfed usng the correlaton coeffcent; however, the tradtonal
More informationChapter 11: Simple Linear Regression and Correlation
Chapter 11: Smple Lnear Regresson and Correlaton 11-1 Emprcal Models 11-2 Smple Lnear Regresson 11-3 Propertes of the Least Squares Estmators 11-4 Hypothess Test n Smple Lnear Regresson 11-4.1 Use of t-tests
More informationEcon Statistical Properties of the OLS estimator. Sanjaya DeSilva
Econ 39 - Statstcal Propertes of the OLS estmator Sanjaya DeSlva September, 008 1 Overvew Recall that the true regresson model s Y = β 0 + β 1 X + u (1) Applyng the OLS method to a sample of data, we estmate
More informationGoodness of fit and Wilks theorem
DRAFT 0.0 Glen Cowan 3 June, 2013 Goodness of ft and Wlks theorem Suppose we model data y wth a lkelhood L(µ) that depends on a set of N parameters µ = (µ 1,...,µ N ). Defne the statstc t µ ln L(µ) L(ˆµ),
More informationx = , so that calculated
Stat 4, secton Sngle Factor ANOVA notes by Tm Plachowsk n chapter 8 we conducted hypothess tests n whch we compared a sngle sample s mean or proporton to some hypotheszed value Chapter 9 expanded ths to
More informationDouble Acceptance Sampling Plan for Time Truncated Life Tests Based on Transmuted Generalized Inverse Weibull Distribution
J. Stat. Appl. Pro. 6, No. 1, 1-6 2017 1 Journal of Statstcs Applcatons & Probablty An Internatonal Journal http://dx.do.org/10.18576/jsap/060101 Double Acceptance Samplng Plan for Tme Truncated Lfe Tests
More informationA note on almost sure behavior of randomly weighted sums of φ-mixing random variables with φ-mixing weights
ACTA ET COMMENTATIONES UNIVERSITATIS TARTUENSIS DE MATHEMATICA Volume 7, Number 2, December 203 Avalable onlne at http://acutm.math.ut.ee A note on almost sure behavor of randomly weghted sums of φ-mxng
More informationComputation of Higher Order Moments from Two Multinomial Overdispersion Likelihood Models
Computaton of Hgher Order Moments from Two Multnomal Overdsperson Lkelhood Models BY J. T. NEWCOMER, N. K. NEERCHAL Department of Mathematcs and Statstcs, Unversty of Maryland, Baltmore County, Baltmore,
More informationOn mutual information estimation for mixed-pair random variables
On mutual nformaton estmaton for mxed-par random varables November 3, 218 Aleksandr Beknazaryan, Xn Dang and Haln Sang 1 Department of Mathematcs, The Unversty of Msssspp, Unversty, MS 38677, USA. E-mal:
More informationStatistics for Economics & Business
Statstcs for Economcs & Busness Smple Lnear Regresson Learnng Objectves In ths chapter, you learn: How to use regresson analyss to predct the value of a dependent varable based on an ndependent varable
More informationNotes on Frequency Estimation in Data Streams
Notes on Frequency Estmaton n Data Streams In (one of) the data streamng model(s), the data s a sequence of arrvals a 1, a 2,..., a m of the form a j = (, v) where s the dentty of the tem and belongs to
More informationThe lower and upper bounds on Perron root of nonnegative irreducible matrices
Journal of Computatonal Appled Mathematcs 217 (2008) 259 267 wwwelsevercom/locate/cam The lower upper bounds on Perron root of nonnegatve rreducble matrces Guang-Xn Huang a,, Feng Yn b,keguo a a College
More informationSee Book Chapter 11 2 nd Edition (Chapter 10 1 st Edition)
Count Data Models See Book Chapter 11 2 nd Edton (Chapter 10 1 st Edton) Count data consst of non-negatve nteger values Examples: number of drver route changes per week, the number of trp departure changes
More informationx i1 =1 for all i (the constant ).
Chapter 5 The Multple Regresson Model Consder an economc model where the dependent varable s a functon of K explanatory varables. The economc model has the form: y = f ( x,x,..., ) xk Approxmate ths by
More informationStatistical inference for generalized Pareto distribution based on progressive Type-II censored data with random removals
Internatonal Journal of Scentfc World, 2 1) 2014) 1-9 c Scence Publshng Corporaton www.scencepubco.com/ndex.php/ijsw do: 10.14419/jsw.v21.1780 Research Paper Statstcal nference for generalzed Pareto dstrbuton
More informationANSWERS. Problem 1. and the moment generating function (mgf) by. defined for any real t. Use this to show that E( U) var( U)
Econ 413 Exam 13 H ANSWERS Settet er nndelt 9 deloppgaver, A,B,C, som alle anbefales å telle lkt for å gøre det ltt lettere å stå. Svar er gtt . Unfortunately, there s a prntng error n the hnt of
More information4 Analysis of Variance (ANOVA) 5 ANOVA. 5.1 Introduction. 5.2 Fixed Effects ANOVA
4 Analyss of Varance (ANOVA) 5 ANOVA 51 Introducton ANOVA ANOVA s a way to estmate and test the means of multple populatons We wll start wth one-way ANOVA If the populatons ncluded n the study are selected
More informationMATH 829: Introduction to Data Mining and Analysis The EM algorithm (part 2)
1/16 MATH 829: Introducton to Data Mnng and Analyss The EM algorthm (part 2) Domnque Gullot Departments of Mathematcal Scences Unversty of Delaware Aprl 20, 2016 Recall 2/16 We are gven ndependent observatons
More informationThe Order Relation and Trace Inequalities for. Hermitian Operators
Internatonal Mathematcal Forum, Vol 3, 08, no, 507-57 HIKARI Ltd, wwwm-hkarcom https://doorg/0988/mf088055 The Order Relaton and Trace Inequaltes for Hermtan Operators Y Huang School of Informaton Scence
More informationUsing T.O.M to Estimate Parameter of distributions that have not Single Exponential Family
IOSR Journal of Mathematcs IOSR-JM) ISSN: 2278-5728. Volume 3, Issue 3 Sep-Oct. 202), PP 44-48 www.osrjournals.org Usng T.O.M to Estmate Parameter of dstrbutons that have not Sngle Exponental Famly Jubran
More informationLinear Approximation with Regularization and Moving Least Squares
Lnear Approxmaton wth Regularzaton and Movng Least Squares Igor Grešovn May 007 Revson 4.6 (Revson : March 004). 5 4 3 0.5 3 3.5 4 Contents: Lnear Fttng...4. Weghted Least Squares n Functon Approxmaton...
More informationCopyright 2017 by Taylor Enterprises, Inc., All Rights Reserved. Adjusted Control Limits for P Charts. Dr. Wayne A. Taylor
Taylor Enterprses, Inc. Control Lmts for P Charts Copyrght 2017 by Taylor Enterprses, Inc., All Rghts Reserved. Control Lmts for P Charts Dr. Wayne A. Taylor Abstract: P charts are used for count data
More informationA new Approach for Solving Linear Ordinary Differential Equations
, ISSN 974-57X (Onlne), ISSN 974-5718 (Prnt), Vol. ; Issue No. 1; Year 14, Copyrght 13-14 by CESER PUBLICATIONS A new Approach for Solvng Lnear Ordnary Dfferental Equatons Fawz Abdelwahd Department of
More informationParameters Estimation of the Modified Weibull Distribution Based on Type I Censored Samples
Appled Mathematcal Scences, Vol. 5, 011, no. 59, 899-917 Parameters Estmaton of the Modfed Webull Dstrbuton Based on Type I Censored Samples Soufane Gasm École Supereure des Scences et Technques de Tuns
More informationInterval Estimation in the Classical Normal Linear Regression Model. 1. Introduction
ECONOMICS 35* -- NOTE 7 ECON 35* -- NOTE 7 Interval Estmaton n the Classcal Normal Lnear Regresson Model Ths note outlnes the basc elements of nterval estmaton n the Classcal Normal Lnear Regresson Model
More informationHere is the rationale: If X and y have a strong positive relationship to one another, then ( x x) will tend to be positive when ( y y)
Secton 1.5 Correlaton In the prevous sectons, we looked at regresson and the value r was a measurement of how much of the varaton n y can be attrbuted to the lnear relatonshp between y and x. In ths secton,
More informationDr. Shalabh Department of Mathematics and Statistics Indian Institute of Technology Kanpur
Analyss of Varance and Desgn of Experment-I MODULE VII LECTURE - 3 ANALYSIS OF COVARIANCE Dr Shalabh Department of Mathematcs and Statstcs Indan Insttute of Technology Kanpur Any scentfc experment s performed
More informationLimited Dependent Variables
Lmted Dependent Varables. What f the left-hand sde varable s not a contnuous thng spread from mnus nfnty to plus nfnty? That s, gven a model = f (, β, ε, where a. s bounded below at zero, such as wages
More informationJoint Statistical Meetings - Biopharmaceutical Section
Iteratve Ch-Square Test for Equvalence of Multple Treatment Groups Te-Hua Ng*, U.S. Food and Drug Admnstraton 1401 Rockvlle Pke, #200S, HFM-217, Rockvlle, MD 20852-1448 Key Words: Equvalence Testng; Actve
More informationComparison of the Population Variance Estimators. of 2-Parameter Exponential Distribution Based on. Multiple Criteria Decision Making Method
Appled Mathematcal Scences, Vol. 7, 0, no. 47, 07-0 HIARI Ltd, www.m-hkar.com Comparson of the Populaton Varance Estmators of -Parameter Exponental Dstrbuton Based on Multple Crtera Decson Makng Method
More informationUSE OF DOUBLE SAMPLING SCHEME IN ESTIMATING THE MEAN OF STRATIFIED POPULATION UNDER NON-RESPONSE
STATISTICA, anno LXXV, n. 4, 015 USE OF DOUBLE SAMPLING SCHEME IN ESTIMATING THE MEAN OF STRATIFIED POPULATION UNDER NON-RESPONSE Manoj K. Chaudhary 1 Department of Statstcs, Banaras Hndu Unversty, Varanas,
More informationEstimation of the Mean of Truncated Exponential Distribution
Journal of Mathematcs and Statstcs 4 (4): 84-88, 008 ISSN 549-644 008 Scence Publcatons Estmaton of the Mean of Truncated Exponental Dstrbuton Fars Muslm Al-Athar Department of Mathematcs, Faculty of Scence,
More informationLecture 4 Hypothesis Testing
Lecture 4 Hypothess Testng We may wsh to test pror hypotheses about the coeffcents we estmate. We can use the estmates to test whether the data rejects our hypothess. An example mght be that we wsh to
More informationImprovement in Estimating the Population Mean Using Exponential Estimator in Simple Random Sampling
Bulletn of Statstcs & Economcs Autumn 009; Volume 3; Number A09; Bull. Stat. Econ. ISSN 0973-70; Copyrght 009 by BSE CESER Improvement n Estmatng the Populaton Mean Usng Eponental Estmator n Smple Random
More informationj) = 1 (note sigma notation) ii. Continuous random variable (e.g. Normal distribution) 1. density function: f ( x) 0 and f ( x) dx = 1
Random varables Measure of central tendences and varablty (means and varances) Jont densty functons and ndependence Measures of assocaton (covarance and correlaton) Interestng result Condtonal dstrbutons
More informationECONOMICS 351*-A Mid-Term Exam -- Fall Term 2000 Page 1 of 13 pages. QUEEN'S UNIVERSITY AT KINGSTON Department of Economics
ECOOMICS 35*-A Md-Term Exam -- Fall Term 000 Page of 3 pages QUEE'S UIVERSITY AT KIGSTO Department of Economcs ECOOMICS 35* - Secton A Introductory Econometrcs Fall Term 000 MID-TERM EAM ASWERS MG Abbott
More informationStat260: Bayesian Modeling and Inference Lecture Date: February 22, Reference Priors
Stat60: Bayesan Modelng and Inference Lecture Date: February, 00 Reference Prors Lecturer: Mchael I. Jordan Scrbe: Steven Troxler and Wayne Lee In ths lecture, we assume that θ R; n hgher-dmensons, reference
More informationExponential Type Product Estimator for Finite Population Mean with Information on Auxiliary Attribute
Avalable at http://pvamu.edu/aam Appl. Appl. Math. ISSN: 193-9466 Vol. 10, Issue 1 (June 015), pp. 106-113 Applcatons and Appled Mathematcs: An Internatonal Journal (AAM) Exponental Tpe Product Estmator
More informationEconomics 130. Lecture 4 Simple Linear Regression Continued
Economcs 130 Lecture 4 Contnued Readngs for Week 4 Text, Chapter and 3. We contnue wth addressng our second ssue + add n how we evaluate these relatonshps: Where do we get data to do ths analyss? How do
More informationNumber of cases Number of factors Number of covariates Number of levels of factor i. Value of the dependent variable for case k
ANOVA Model and Matrx Computatons Notaton The followng notaton s used throughout ths chapter unless otherwse stated: N F CN Y Z j w W Number of cases Number of factors Number of covarates Number of levels
More informationIntroduction to Regression
Introducton to Regresson Dr Tom Ilvento Department of Food and Resource Economcs Overvew The last part of the course wll focus on Regresson Analyss Ths s one of the more powerful statstcal technques Provdes
More informationUNIVERSITY OF TORONTO Faculty of Arts and Science. December 2005 Examinations STA437H1F/STA1005HF. Duration - 3 hours
UNIVERSITY OF TORONTO Faculty of Arts and Scence December 005 Examnatons STA47HF/STA005HF Duraton - hours AIDS ALLOWED: (to be suppled by the student) Non-programmable calculator One handwrtten 8.5'' x
More informationDiscussion of Extensions of the Gauss-Markov Theorem to the Case of Stochastic Regression Coefficients Ed Stanek
Dscusson of Extensons of the Gauss-arkov Theorem to the Case of Stochastc Regresson Coeffcents Ed Stanek Introducton Pfeffermann (984 dscusses extensons to the Gauss-arkov Theorem n settngs where regresson
More informationEconometrics of Panel Data
Econometrcs of Panel Data Jakub Mućk Meetng # 8 Jakub Mućk Econometrcs of Panel Data Meetng # 8 1 / 17 Outlne 1 Heterogenety n the slope coeffcents 2 Seemngly Unrelated Regresson (SUR) 3 Swamy s random
More information2016 Wiley. Study Session 2: Ethical and Professional Standards Application
6 Wley Study Sesson : Ethcal and Professonal Standards Applcaton LESSON : CORRECTION ANALYSIS Readng 9: Correlaton and Regresson LOS 9a: Calculate and nterpret a sample covarance and a sample correlaton
More informationSupplementary material: Margin based PU Learning. Matrix Concentration Inequalities
Supplementary materal: Margn based PU Learnng We gve the complete proofs of Theorem and n Secton We frst ntroduce the well-known concentraton nequalty, so the covarance estmator can be bounded Then we
More informationAn (almost) unbiased estimator for the S-Gini index
An (almost unbased estmator for the S-Gn ndex Thomas Demuynck February 25, 2009 Abstract Ths note provdes an unbased estmator for the absolute S-Gn and an almost unbased estmator for the relatve S-Gn for
More informationNegative Binomial Regression
STATGRAPHICS Rev. 9/16/2013 Negatve Bnomal Regresson Summary... 1 Data Input... 3 Statstcal Model... 3 Analyss Summary... 4 Analyss Optons... 7 Plot of Ftted Model... 8 Observed Versus Predcted... 10 Predctons...
More informationChapter 2 - The Simple Linear Regression Model S =0. e i is a random error. S β2 β. This is a minimization problem. Solution is a calculus exercise.
Chapter - The Smple Lnear Regresson Model The lnear regresson equaton s: where y + = β + β e for =,..., y and are observable varables e s a random error How can an estmaton rule be constructed for the
More information/ n ) are compared. The logic is: if the two
STAT C141, Sprng 2005 Lecture 13 Two sample tests One sample tests: examples of goodness of ft tests, where we are testng whether our data supports predctons. Two sample tests: called as tests of ndependence
More informationHowever, since P is a symmetric idempotent matrix, of P are either 0 or 1 [Eigen-values
Fall 007 Soluton to Mdterm Examnaton STAT 7 Dr. Goel. [0 ponts] For the general lnear model = X + ε, wth uncorrelated errors havng mean zero and varance σ, suppose that the desgn matrx X s not necessarly
More informationAPPENDIX A Some Linear Algebra
APPENDIX A Some Lnear Algebra The collecton of m, n matrces A.1 Matrces a 1,1,..., a 1,n A = a m,1,..., a m,n wth real elements a,j s denoted by R m,n. If n = 1 then A s called a column vector. Smlarly,
More informationBasically, if you have a dummy dependent variable you will be estimating a probability.
ECON 497: Lecture Notes 13 Page 1 of 1 Metropoltan State Unversty ECON 497: Research and Forecastng Lecture Notes 13 Dummy Dependent Varable Technques Studenmund Chapter 13 Bascally, f you have a dummy
More informationNUMERICAL DIFFERENTIATION
NUMERICAL DIFFERENTIATION 1 Introducton Dfferentaton s a method to compute the rate at whch a dependent output y changes wth respect to the change n the ndependent nput x. Ths rate of change s called the
More informationLecture 3: Probability Distributions
Lecture 3: Probablty Dstrbutons Random Varables Let us begn by defnng a sample space as a set of outcomes from an experment. We denote ths by S. A random varable s a functon whch maps outcomes nto the
More informationSTAT 3008 Applied Regression Analysis
STAT 3008 Appled Regresson Analyss Tutoral : Smple Lnear Regresson LAI Chun He Department of Statstcs, The Chnese Unversty of Hong Kong 1 Model Assumpton To quantfy the relatonshp between two factors,
More informationOutline. Zero Conditional mean. I. Motivation. 3. Multiple Regression Analysis: Estimation. Read Wooldridge (2013), Chapter 3.
Outlne 3. Multple Regresson Analyss: Estmaton I. Motvaton II. Mechancs and Interpretaton of OLS Read Wooldrdge (013), Chapter 3. III. Expected Values of the OLS IV. Varances of the OLS V. The Gauss Markov
More informationChapter 5 Multilevel Models
Chapter 5 Multlevel Models 5.1 Cross-sectonal multlevel models 5.1.1 Two-level models 5.1.2 Multple level models 5.1.3 Multple level modelng n other felds 5.2 Longtudnal multlevel models 5.2.1 Two-level
More informationDr. Shalabh Department of Mathematics and Statistics Indian Institute of Technology Kanpur
Analyss of Varance and Desgn of Exerments-I MODULE III LECTURE - 2 EXPERIMENTAL DESIGN MODELS Dr. Shalabh Deartment of Mathematcs and Statstcs Indan Insttute of Technology Kanur 2 We consder the models
More informationInner Product. Euclidean Space. Orthonormal Basis. Orthogonal
Inner Product Defnton 1 () A Eucldean space s a fnte-dmensonal vector space over the reals R, wth an nner product,. Defnton 2 (Inner Product) An nner product, on a real vector space X s a symmetrc, blnear,
More informationPrimer on High-Order Moment Estimators
Prmer on Hgh-Order Moment Estmators Ton M. Whted July 2007 The Errors-n-Varables Model We wll start wth the classcal EIV for one msmeasured regressor. The general case s n Erckson and Whted Econometrc
More informationDr. Shalabh Department of Mathematics and Statistics Indian Institute of Technology Kanpur
Analyss of Varance and Desgn of Experment-I MODULE VIII LECTURE - 34 ANALYSIS OF VARIANCE IN RANDOM-EFFECTS MODEL AND MIXED-EFFECTS EFFECTS MODEL Dr Shalabh Department of Mathematcs and Statstcs Indan
More informationExercises of Chapter 2
Exercses of Chapter Chuang-Cheh Ln Department of Computer Scence and Informaton Engneerng, Natonal Chung Cheng Unversty, Mng-Hsung, Chay 61, Tawan. Exercse.6. Suppose that we ndependently roll two standard
More informationENG 8801/ Special Topics in Computer Engineering: Pattern Recognition. Memorial University of Newfoundland Pattern Recognition
EG 880/988 - Specal opcs n Computer Engneerng: Pattern Recognton Memoral Unversty of ewfoundland Pattern Recognton Lecture 7 May 3, 006 http://wwwengrmunca/~charlesr Offce Hours: uesdays hursdays 8:30-9:30
More informationHomework 9 STAT 530/J530 November 22 nd, 2005
Homework 9 STAT 530/J530 November 22 nd, 2005 Instructor: Bran Habng 1) Dstrbuton Q-Q plot Boxplot Heavy Taled Lght Taled Normal Skewed Rght Department of Statstcs LeConte 203 ch-square dstrbuton, Telephone:
More informationMaximum Likelihood Estimation of Binary Dependent Variables Models: Probit and Logit. 1. General Formulation of Binary Dependent Variables Models
ECO 452 -- OE 4: Probt and Logt Models ECO 452 -- OE 4 Maxmum Lkelhood Estmaton of Bnary Dependent Varables Models: Probt and Logt hs note demonstrates how to formulate bnary dependent varables models
More informationChapter 13: Multiple Regression
Chapter 13: Multple Regresson 13.1 Developng the multple-regresson Model The general model can be descrbed as: It smplfes for two ndependent varables: The sample ft parameter b 0, b 1, and b are used to
More information1. Inference on Regression Parameters a. Finding Mean, s.d and covariance amongst estimates. 2. Confidence Intervals and Working Hotelling Bands
Content. Inference on Regresson Parameters a. Fndng Mean, s.d and covarance amongst estmates.. Confdence Intervals and Workng Hotellng Bands 3. Cochran s Theorem 4. General Lnear Testng 5. Measures of
More informationBayesian predictive Configural Frequency Analysis
Psychologcal Test and Assessment Modelng, Volume 54, 2012 (3), 285-292 Bayesan predctve Confgural Frequency Analyss Eduardo Gutérrez-Peña 1 Abstract Confgural Frequency Analyss s a method for cell-wse
More informationOutline. Bayesian Networks: Maximum Likelihood Estimation and Tree Structure Learning. Our Model and Data. Outline
Outlne Bayesan Networks: Maxmum Lkelhood Estmaton and Tree Structure Learnng Huzhen Yu janey.yu@cs.helsnk.f Dept. Computer Scence, Unv. of Helsnk Probablstc Models, Sprng, 200 Notces: I corrected a number
More informationA Comparative Study for Estimation Parameters in Panel Data Model
A Comparatve Study for Estmaton Parameters n Panel Data Model Ahmed H. Youssef and Mohamed R. Abonazel hs paper examnes the panel data models when the regresson coeffcents are fxed random and mxed and
More informatione i is a random error
Chapter - The Smple Lnear Regresson Model The lnear regresson equaton s: where + β + β e for,..., and are observable varables e s a random error How can an estmaton rule be constructed for the unknown
More informationThe Ordinary Least Squares (OLS) Estimator
The Ordnary Least Squares (OLS) Estmator 1 Regresson Analyss Regresson Analyss: a statstcal technque for nvestgatng and modelng the relatonshp between varables. Applcatons: Engneerng, the physcal and chemcal
More informationStatistics for Managers Using Microsoft Excel/SPSS Chapter 13 The Simple Linear Regression Model and Correlation
Statstcs for Managers Usng Mcrosoft Excel/SPSS Chapter 13 The Smple Lnear Regresson Model and Correlaton 1999 Prentce-Hall, Inc. Chap. 13-1 Chapter Topcs Types of Regresson Models Determnng the Smple Lnear
More informationOn the Influential Points in the Functional Circular Relationship Models
On the Influental Ponts n the Functonal Crcular Relatonshp Models Department of Mathematcs, Faculty of Scence Al-Azhar Unversty-Gaza, Gaza, Palestne alzad33@yahoo.com Abstract If the nterest s to calbrate
More informationRandom Partitions of Samples
Random Parttons of Samples Klaus Th. Hess Insttut für Mathematsche Stochastk Technsche Unverstät Dresden Abstract In the present paper we construct a decomposton of a sample nto a fnte number of subsamples
More informationEfficient nonresponse weighting adjustment using estimated response probability
Effcent nonresponse weghtng adjustment usng estmated response probablty Jae Kwang Km Department of Appled Statstcs, Yonse Unversty, Seoul, 120-749, KOREA Key Words: Regresson estmator, Propensty score,
More informationStat 642, Lecture notes for 01/27/ d i = 1 t. n i t nj. n j
Stat 642, Lecture notes for 01/27/05 18 Rate Standardzaton Contnued: Note that f T n t where T s the cumulatve follow-up tme and n s the number of subjects at rsk at the mdpont or nterval, and d s the
More information4.3 Poisson Regression
of teratvely reweghted least squares regressons (the IRLS algorthm). We do wthout gvng further detals, but nstead focus on the practcal applcaton. > glm(survval~log(weght)+age, famly="bnomal", data=baby)
More informationANOMALIES OF THE MAGNITUDE OF THE BIAS OF THE MAXIMUM LIKELIHOOD ESTIMATOR OF THE REGRESSION SLOPE
P a g e ANOMALIES OF THE MAGNITUDE OF THE BIAS OF THE MAXIMUM LIKELIHOOD ESTIMATOR OF THE REGRESSION SLOPE Darmud O Drscoll ¹, Donald E. Ramrez ² ¹ Head of Department of Mathematcs and Computer Studes
More informationLINEAR REGRESSION ANALYSIS. MODULE IX Lecture Multicollinearity
LINEAR REGRESSION ANALYSIS MODULE IX Lecture - 31 Multcollnearty Dr. Shalabh Department of Mathematcs and Statstcs Indan Insttute of Technology Kanpur 6. Rdge regresson The OLSE s the best lnear unbased
More informationSupporting Information
Supportng Informaton The neural network f n Eq. 1 s gven by: f x l = ReLU W atom x l + b atom, 2 where ReLU s the element-wse rectfed lnear unt, 21.e., ReLUx = max0, x, W atom R d d s the weght matrx to
More informationCOMPARISON OF SOME RELIABILITY CHARACTERISTICS BETWEEN REDUNDANT SYSTEMS REQUIRING SUPPORTING UNITS FOR THEIR OPERATIONS
Avalable onlne at http://sck.org J. Math. Comput. Sc. 3 (3), No., 6-3 ISSN: 97-537 COMPARISON OF SOME RELIABILITY CHARACTERISTICS BETWEEN REDUNDANT SYSTEMS REQUIRING SUPPORTING UNITS FOR THEIR OPERATIONS
More information2E Pattern Recognition Solutions to Introduction to Pattern Recognition, Chapter 2: Bayesian pattern classification
E395 - Pattern Recognton Solutons to Introducton to Pattern Recognton, Chapter : Bayesan pattern classfcaton Preface Ths document s a soluton manual for selected exercses from Introducton to Pattern Recognton
More informationSampling Theory MODULE V LECTURE - 17 RATIO AND PRODUCT METHODS OF ESTIMATION
Samplng Theory MODULE V LECTURE - 7 RATIO AND PRODUCT METHODS OF ESTIMATION DR. SHALABH DEPARTMENT OF MATHEMATICS AND STATISTICS INDIAN INSTITUTE OF TECHNOLOG KANPUR Propertes of separate rato estmator:
More informationLectures - Week 4 Matrix norms, Conditioning, Vector Spaces, Linear Independence, Spanning sets and Basis, Null space and Range of a Matrix
Lectures - Week 4 Matrx norms, Condtonng, Vector Spaces, Lnear Independence, Spannng sets and Bass, Null space and Range of a Matrx Matrx Norms Now we turn to assocatng a number to each matrx. We could
More informationChapter 3 Describing Data Using Numerical Measures
Chapter 3 Student Lecture Notes 3-1 Chapter 3 Descrbng Data Usng Numercal Measures Fall 2006 Fundamentals of Busness Statstcs 1 Chapter Goals To establsh the usefulness of summary measures of data. The
More informationOnline Appendix to: Axiomatization and measurement of Quasi-hyperbolic Discounting
Onlne Appendx to: Axomatzaton and measurement of Quas-hyperbolc Dscountng José Lus Montel Olea Tomasz Strzaleck 1 Sample Selecton As dscussed before our ntal sample conssts of two groups of subjects. Group
More informationStat 543 Exam 2 Spring 2016
Stat 543 Exam 2 Sprng 206 I have nether gven nor receved unauthorzed assstance on ths exam. Name Sgned Date Name Prnted Ths Exam conssts of questons. Do at least 0 of the parts of the man exam. I wll score
More informationLOW BIAS INTEGRATED PATH ESTIMATORS. James M. Calvin
Proceedngs of the 007 Wnter Smulaton Conference S G Henderson, B Bller, M-H Hseh, J Shortle, J D Tew, and R R Barton, eds LOW BIAS INTEGRATED PATH ESTIMATORS James M Calvn Department of Computer Scence
More information