The Nottingham eprints service makes this work by researchers of the University of Nottingham available open access under the following conditions.

Size: px
Start display at page:

Download "The Nottingham eprints service makes this work by researchers of the University of Nottingham available open access under the following conditions."

Transcription

1 Harvey, David I. and Leybourne, Stehen J. and Taylor, A.M. Robert (04) On infimum Dickey Fuller unit root tests allowing for a trend break under the null. Comutational Statistics & Data Analysis, ISSN Access from the University of Nottingham reository: htt://erints.nottingham.ac.uk/33//zar.df Coyright and reuse: The Nottingham eprints service makes this work by researchers of the University of Nottingham available oen access under the following conditions. This article is made available under the University of Nottingham End User licence and may be reused according to the conditions of the licence. For more details see: htt://erints.nottingham.ac.uk/end_user_agreement.df A note on versions: The version resented here may differ from the ublished version or from the version of record. If you wish to cite this item you are advised to consult the ublisher s version. Please see the reository url above for details on accessing the ublished version and note that access may require a subscrition. For more information, lease contact erints@nottingham.ac.uk

2 On Infimum Dickey-Fuller Unit Root Tests Allowing for a Trend Break Under The Null David I. Harvey, Stehen J. Leybourne and A.M. Robert Taylor Granger Centre for Time Series Econometrics and School of Economics, University of Nottingham October 0 Abstract Trend breaks aear to be revalent in macroeconomic time series. Consequently, to avoid the catastrohic imact that unmodelled trend breaks have on ower it is standard emirical ractice to emloy unit root tests which allow for such effects. A oularly alied aroach is the infimum ADF-tye test. Its aeal has endured with ractitioners desite results which show that the infimum ADF statistic diverges to as the samle size diverges, with the consequence that the test has an asymtotic size of unity when a break in trend is resent under the unit root null hyothesis. The result for additive outlier-tye breaks in trend (but not intercet) is refined and shows that divergence to occurs only when the true break fraction is smaller than /3. An alternative testing strategy based on the maximum of the original infimum statistic and the corresonding statistic constructed using the time-reversed samle data is considered. Keywords: Unit root test; trend break; Minimum Dickey-Fuller test. Address corresondence to: David Harvey, School of Economics, University of Nottingham, Nottingham NG7 RD, UK. Tel.: +44 (0) Fax: +44 (0) dave.harvey@nottingham.ac.uk

3 Introduction Macroeconomic series aear to often be characterized by broken trend functions; see, inter alia, Stock and Watson (99,999,005) and Perron and Zhu (005). In a seminal aer, Perron (989) shows that failure to account for trend breaks resent in the data results in unit root tests with zero ower, even asymtotically. Consequently, when testing for a unit root it has become a matter of regular ractice to allow for this kind of deterministic structural change. While Perron (989) initially treated the location of the break as known, subsequent aroaches have focused on the case where the break location is unknown and is chosen via a data-deendent method; see, inter alia, Zivot and Andrews (99) [ZA], Banerjee et al. (99) and Perron (997); see also Pitarakis (0). Of these endogenised aroaches, the testing methodology roosed by ZA has been widely used by ractitioners (for a recent examle, see Alexeev and Maynard, 0). The aroach suggested by ZA is to calculate the ADF t-ratio-tye statistic of Perron (989) for all candidate break oints within a trimmed range and to then form a test based on the infimum (most negative) of this sequence of statistics. This infimum test is simle to comute and, by selecting the statistic within the sequence which gives most weight to the alternative, follows an established aroach to such roblems in econometrics. A significant drawback with the infimum aroach, however, is that it is redicated on the maintained hyothesis that no break in trend occurs under the unit root null hyothesis. This assumtion needs to be made in order for the infimum statistic to have a ivotal limiting null distribution. Investigating what haens when this maintained assumtion does not hold, results resented in Vogelsang and Perron (998) show that, for both sudden additive outlier () breaks and slowly evolving innovational outlier (IO) breaks, when a trend break of fixed non-zero magnitude occurs under the unit root null, so the infimum statistics diverge to as the samle size diverges and, hence, cause the tests to have asymtotic size of unity. In this aer we revisit this issue, focusing on -tye breaks in the trend (but not intercet). Our rimary contribution is to refine the theoretical results given in Vogelsang and Perron (998), showing that the divergence of the infimum statistic to occurs only when the true break fraction, τ 0 say, is smaller than /3. We also briefly consider an alternative testing strategy based on the maximum of the original infimum statistic and the corresonding statistic constructed using the time-reversed samle data. We find that such an aroach aears to offer considerable imrovements in finite samle size control relative to the original test, while retaining attractive ower roerties. In the following denotes the integer art of its argument, and denote weak convergence and convergence in robability, resectively, in each case as the samle size diverges to +, x := y ( x =: y ) indicates that x is defined by y (y is defined by x), and ( ) denotes the indicator function.

4 The Model and the Infimum Unit Root Test We consider a univariate time series {y t } generated by the DGP, y t = µ + βt + γdt t (τ 0 ) + u t, t =,..., T, () u t = ρu t + ε t, t =,..., T () where, for a generic fraction τ, DT t (τ) := (t > τt )(t τt ) in (), and τ 0 is an (unknown) utative trend break fraction, with associated break magnitude γ. The break fraction is assumed to be such that τ 0 Λ, where Λ := [τ L, τ U ] with 0 < τ L < τ U < ; the fractions τ L and τ U reresenting trimming arameters. In (), {u t } is an unobserved mean zero stochastic rocess, initialised such that u = o (T / ). Also, for simlicity of exosition, we will assume that ε t in () is an indeendent and identically distributed sequence with mean zero, variance σ ε and finite fourth moment. The theoretical results given in the aer would continue to hold under a more general weak deendence assumtion rovided the Dickey-Fuller-tye unit root regression in (3) below was augmented with k lags of the deendent variable and where k satisfies the usual condition that /k + k 3 /T 0 as T. We examine the roblem of testing the unit root null hyothesis H 0 : ρ =, against the alternative, H : ρ <, without assuming knowledge of where, or indeed if, the trend break occurs in the DGP. Let û t denote the residual from fitting the OLS regression of y t on z t := [, t, DT t (τ)] (we suress the deendence of û t and any associated OLS estimators on τ for convenience of notation), i.e. û t := y t ˆµ ˆβt ˆγDT t (τ) and let tˆφ (τ) denote the Dickey-Fuller t-ratio for testing φ = 0 in the fitted OLS regression û t = ˆφû t + ˆε t (3) that is tˆφ (τ) := T t= û tû t ˆσ ε T t= û t with ˆσ ε := (T ) T t= ˆε t. Then the infimum ZA-tye rocedure we consider is based on the statistic ZA := inf τ Λ tˆφ (τ). 3 Limiting Behaviour of ZA In order to derive the large samle behaviour of the ZA statistic we must first evaluate the limiting behaviour of the tˆφ (τ) statistic for τ τ 0. This is rovided in Theorem. Theorem Let y t be generated according to () and (), with γ = κσ ε, κ 0. Then for τ τ 0 T / tˆφ (τ) κ {N(, τ 0, τ) N(0, τ 0, τ) } κ { + κ M(τ 0, τ)} 0 N(r, τ 0, τ) dr

5 with N(r, τ 0, τ) := I r τ 0 (r τ 0 ) P P r I r τ P 3 (r τ), M(τ 0, τ) := L(τ 0, τ) {N(, τ 0, τ) N(0, τ 0, τ) } 0 N(r, τ, 0, τ) dr L(τ 0, τ) := ( τ 0 ) + P + ( τ)p3 ( τ 0 )P ( τ 0 I τ τ 0 (τ τ 0 ))P 3 + ( τ)p P 3 and P P P 3 := 3 ( τ) ( τ) (+τ) ( τ) ( τ) (+τ) ( τ) 3 3 ( τ 0 ) ( τ 0 ) (+τ 0 ) { τ 0 I τ τ 0 (τ τ 0 )} {+τ 0 3τ+4I τ τ 0 (τ τ 0 )} where I y x := (y > x). Notice that when τ = τ 0, we find that [P, P, P 3 ] = [0, 0, ] and N(r, τ 0, τ 0 ) = 0; consequently, the limit of T / tˆφ (τ 0 ) is undefined. We know, however, from Kim and Perron (009) that tˆφ (τ) has a well-defined limit distribution under H 0 when τ is within an o(t / ) neighbourhood of τ 0, thus tˆφ (τ) = O () here, and we will therefore consider the limit of T / tˆφ (τ 0 ) to be zero. For τ τ 0, Theorem imlies that tˆφ (τ) = O (T / ). Now since ZA takes the minimum value of tˆφ (τ) across all τ Λ, the ertinent issue is the sign of the limit of T / tˆφ (τ). More secifically, if, given a break at time τ 0, T / tˆφ (τ) is ositive for all τ Λ, then, since tˆφ (τ) + for all τ τ 0, ZA would not be minimised over this roblem region, but rather for a value of τ within a shrinking neighbourhood of τ 0. On the other hand, if T / tˆφ (τ) is negative for any τ Λ, then tˆφ (τ) for some τ τ 0, and consequently ZA also, resulting in unit asymtotic size. We now therefore examine the sign of the limit of T / tˆφ (τ) as a function of τ 0 and τ. The sign of the limit of T / tˆφ (τ) is determined by the sign of N(, τ 0, τ) N(0, τ 0, τ) as the other terms in the limit are unambiguously ositive (as is clear from the Proof of Theorem, + κ M(τ 0, τ) is the limit of ˆσ ε/σ ε and is therefore ositive). Next note that we can write N(0, τ 0, τ) = P, N(, τ 0, τ) = ( τ 0 ) P P P 3 ( τ) and so N(, τ 0, τ) N(0, τ 0, τ) = {( τ 0 ) P P P 3 ( τ)} P. First consider the case where τ < τ 0. Here, we find (uon simlification) P P P 3 = ( τ 0 ) (τ 0 τ) ( τ) 3( τ 0) (τ 0 τ) τ( τ) ( τ 0 ) (3τ 0 τ ττ 0 ) τ( τ) 3 yielding N(, τ 0, τ) N(0, τ 0, τ) = j τ,τ 0 h τ,τ 0 (4) 3

6 where j τ,τ 0 := ( τ 0 ) (τ 0 τ) ( τ + τ 0 τ) /4 ( τ) 4, h τ,τ 0 := τ 0 ( τ) ( τ 0 ) with j τ,τ 0 always ositive when τ < τ 0. Now, the function h τ,τ 0 is, for a given τ 0, monotonically decreasing in τ, and since τ L τ < τ 0 it is bounded by [τ 0 ( τ L ) ( τ 0 ), (τ 0 )( τ 0 )). We then find that For τ 0 < /3 h τ,τ 0 < 0 for all τ L τ < τ 0, { < 0 for τ L τ < 3τ 0 τ For /3 τ 0 < / h 0 τ,τ 0, 0 for 3τ 0 τ 0 τ < τ 0 For τ 0 / h τ,τ 0 > 0 for all τ L τ < τ 0. Next, when τ > τ 0 we have P P P 3 = τ 0 (τ τ 0 )( τ+τ 0 +ττ 0 ) τ (τ τ 0 )(τ τ 0 +ττ 0 3ττ 0 ) τ 3 τ 0 ( 3τ+τ 0+ττ 0 ) τ 3 (τ ) giving N(, τ 0, τ) N(0, τ 0, τ) = k τ,τ 0 l τ,τ 0 (5) where k τ,τ 0 := τ 0 (τ τ 0 ) (τ τ 0 ) /4τ 4, l τ,τ 0 := τ 0 τ( τ 0 ) with k τ,τ 0 always ositive when τ > τ 0. The function l τ,τ 0 is, for a given τ 0, monotonically decreasing in τ and since τ 0 < τ τ U it is bounded by (τ 0 (τ 0 ), τ 0 τ U ( τ 0 )]. Then, For τ 0 < / l τ,τ 0 < 0 for all τ 0 < τ τ U, < 0 for τ 0 τ < τ 0 ( τ For / τ 0 < /3 l 0 ) τ,τ 0 τ 0 for 0 ( τ 0 ) τ τ U For τ 0 /3 l τ,τ 0 > 0 for all τ 0 < τ τ U., Drawing on the above results for (4) when τ < τ 0 and (5) when τ > τ 0, we can then write For τ 0 < /3 lim(t / tˆφ (τ)) < 0 for all τ L τ τ U, { < 0 for some τ For /3 τ 0 < /3 lim(t / L τ τ U tˆφ (τ)), 0 for some τ L τ τ U For τ 0 /3 lim(t / tˆφ (τ)) > 0 for all τ L τ τ U. 4

7 To further illustrate the behaviour of T / tˆφ (τ), Figure dislays the regions in (τ, τ 0 ) sace where lim(t / tˆφ (τ)) is ositive/negative. Translating this behaviour into that of ZA, which locates the minimum of tˆφ (τ) across all τ Λ, we consequently obtain that { τ 0 < /3 ZA O () τ 0 /3. We therefore find that, under H 0, ZA will suriously reject with robability aroaching one in the limit rovided the true break fraction τ 0 lies below /3. It will, however, not suriously reject with robability one in the limit if τ 0 is /3 or above. It is this second finding that refines the result resented in Vogelsang and Perron (998), since the finding that surious rejections of the null occur with robability one in the limit is here shown to deend on where the true break is located. (Note that, in contrast, the innovational outlier version of the statistic diverges to for all τ 0, as stated by Vogelsang and Perron (998).) Of course, in an emirical setting where uncertainty exists as to the resence or location of a break, one is unlikely to have any confidence that the utative break will lie in the region τ 0 [/3, τ U ], and so the fundamental roblem raised by Vogelsang and Perron (998) of otentially serious over-sizing ersists in the ractical alication of ZA. lim(t / tˆφ(τ)) < 0 lim(t / tˆφ(τ)) > 0 Figure. Sign of lim(t /tˆφ(τ)) 4 Use of Time-Reversed Data In order to consider how the asymtotic over-size roblem region of a ZA -tye test might be reduced, we now consider the following imrovisation. If we time-reverse the data, i.e. consider {y T t+ } T t= in lace of {y t } T t=, then any break aearing in the first half of the original samle {y t} is translated into one occurring in the second half of {y T t+ }. Thus alication of ZA to {y T t+ }, which we denote by ZA, will deliver a test which does not suriously reject in the limit with robability one when a break occurs in the first third of {y t }. Of course, since in ractice we have no information regarding 5

8 the location of a break, ZA does not achieve robustness, since here a break in the last two-thirds of {y t } would cause surious rejection of the null by ZA. However, if we consider the maximum of ZA and ZA (cf. Leybourne, 995, in the context of unit root testing without allowance for a break in trend); that is, ZA max := max(za, ZA ) the ZA roblem of suriously rejecting the null with robability aroaching one when τ 0 [τ L, /3] restricted to the region τ 0 (/3, /3). Under H 0 and for the case γ = 0, it is straight- is for ZA max forward to show that where ZA max ( ) max inf Z(τ), inf Z (τ) τ Λ τ Λ Z(τ) := K(, τ) K(0, τ), Z (τ) := K (, τ) K (0, τ) 0 K(r, τ) dr 0 K (r, τ) dr with K(r, τ) and K (r, τ) the continuous time residual rocesses from the rojections of W (r), and W ( r), resectively, onto the sace sanned by {, r, (r τ)i r τ }, where W (r) is the Brownian motion rocess defined by T / rt t= ε t σ ε W (r). Table reorts asymtotic nominal 0.0, 0.05 and 0.0 level critical values for ZA max for a selection of commonly used trimming arameters. The critical values were obtained by direct simulation of (), aroximating the Wiener rocesses in the limiting functionals using N IID(0, ) random variates, with the integrals aroximated by normalized sums of,000 stes. Unreorted simulations, available from the authors on request, suggest that (i) ZA max dislays good finite samle size control for all τ 0 values, unless a large magnitude break occurs at τ 0 (/3, /3) in an extremely large samle, and (ii) ZA max has attractive ower roerties under H, regardless of whether or not a break is actually resent. Given that substantial over-size is only seen to occur for series that are unreresentative of those encountered in tyical economic alications, a ragmatic case could be made for using ZA max, although it is difficult to fully justify such an aroach, given that the test still has asymtotic size aroaching one when τ 0 (/3, /3). () Table. Asymtotic ξ-level critical values for the ZA max test [τ L, τ U ] ξ = 0.0 ξ = 0.05 ξ = 0.0 [0.05, 0.95] [0.0, 0.90] [0.5, 0.85] [0.0, 0.80] Acknowledgment Financial suort rovided by the Economic and Social Research Council of the United Kingdom under research grant RES is gratefully acknowledged by the authors.

9 References Alexeev, V. and Maynard, A. (0). Localized level crossing random walk test robust to the resence of structural breaks. Comutational Statistics and Data Analysis 5, Banerjee, A., Lumsdaine, R. and Stock, J. (99). Recursive and sequential tests of the unit root and trend break hyotheses: theory and international evidence. Journal of Business and Economics Statistics 0, Kim, D. and Perron, P. (009). Unit root tests allowing for a break in the trend function at an unknown time under both the null and alternative hyotheses. Journal of Econometrics 48, -3. Leybourne, S.J. (995). Testing for unit roots using forward and reverse Dickey-Fuller regressions. Oxford Bulletin of Economics and Statistics 57, Perron, P. (989). The great crash, the oil rice shock, and the unit root hyothesis. Econometrica 57, Perron, P. (997). Further evidence of breaking trend functions in macroeconomic variables. Journal of Econometrics 80, Perron, P. and Zhu, X. (005). Structural breaks with deterministic and stochastic trends. Journal of Econometrics 9, 5-9. Pitarakis, J.-Y. (0). A joint test for structural stability and a unit root in autoregressions. Comutational Statistics and Data Analysis, forthcoming. Stock, J.H. and Watson, M.W. (99). Evidence on structural instability in macroeconomic time series relations. Journal of Business and Economic Statistics 4, -30. Stock, J.H. and Watson, M.W. (999). A comarison of linear and nonlinear univariate models for forecasting macroeconomic time series. Engle, R.F. and White, H. (eds.), Cointegration, Causality and Forecasting: A Festschrift in Honour of Clive W.J. Granger, Oxford: Oxford University Press, -44. Stock, J. and Watson, M.W. (005). Imlications of dynamic factor analysis for VAR models, NBER Working Paer #47. Vogelsang, T.J. and Perron, P. (998). Additional tests for a unit root allowing the ossibility of breaks in the trend function. International Economic Review 39, Zivot, E. and Andrews, D.W.K. (99). Further evidence on the great crash, the oil-rice shock, and the unit-root hyothesis. Journal of Business and Economic Statistics 0,

10 Aendix: Proof of Theorem In what follows we can set µ = β = 0 without loss of generality. When τ τ 0 we have T T t= y t = T T t= u t + κσ ε T T t=τ 0 T + (t τ 0T ) κσ ε ( τ 0 ) /, T 3 T t= ty t = T 3 T t= tu t + κσ ε T 3 T t=τ 0 T + t(t τ 0T ) κσ ε ( τ 0 ) ( + τ 0 ) /, T 3 T t=τt + (t τt )y t = T 3 T t=τt + (t τt )u t + κσ ε T 3 T t=τt + (t τt )DT t(τ 0 ) κσ ε [{ τ 0 I τ τ 0 (τ τ 0 )} { + τ 0 3τ + 4I τ τ 0 (τ τ 0 )}]/. Examining the limit behaviour of the residuals û t we obtain T ˆµ ˆβ = ˆγ giving We also require T T T T t= t T T t=τt + (t τt ) T T t= t T 3 T t= t T 3 T t=τt + t(t τt ) (t τt ) T 3 T t(t τt ) T (t τt ) t=τt + T T t= y t T 3 T t= ty t T 3 T t=τt + (t τt )y t 3 ( τ) ( τ) (+τ) = : κσ ε P P P 3 ( τ) ( τ) (+τ) ( τ) 3 3 t=τt + 3 T t=τt + κσ ε( τ 0 ) κσ ε( τ 0 ) (+τ 0 ) κσ ε{ τ 0 I τ τ 0 (τ τ 0 )} {+τ 0 3τ+4I τ τ 0 (τ τ 0 )} T û rt = T y rt T ˆµ T ˆβ rt T ˆγI r τ ( rt τt ) = T u rt + κσ ε I r τ 0 (r τ 0 ) T ˆµ ˆβr ˆγI r τ (r τ) û t = y t ˆβ ˆγDU t (τ) κσ ε {I r τ 0 (r τ 0 ) P P r I r τ P 3 (r τ)} =: κσ ε N(r, τ 0, τ). = κσ ε DU t (τ 0 ) + u t ˆβ ˆγDU t (τ) T t= û t = κ σ ε(t τ 0 T ) + T t= u t + (T )ˆβ + ˆγ (T τt ) T T t= û t +κσ ε T t=τ 0 T + u t κσ ε (T τ 0 T )ˆβ κσ εˆγ T t= DU t(τ 0 )DU t (τ) ˆβ T t= u t ˆγ T t=τt + u t + ˆβˆγ(T τt ) σ ε[ + κ {( τ 0 ) + P + ( τ)p3 ( τ 0 )P ( τ 0 I τ τ 0 (τ τ 0 ))P 3 +( τ)p P 3 }] =: σ ε{ + κ L(τ 0, τ)} 8

11 and T ˆφ = T û T T û T T t= û t T 3 T t= û t N(, τ 0, τ) N(0, τ 0, τ) 0 N(r, τ 0, τ) dr. Now we can establish the limit of ˆσ ε ˆσ ε = T T t= ˆε t = T T t= û t + T ˆφ T 3 T t= û t T ˆφT T t= û tû t [ ] σ ε{ + κ N(, τ 0, τ) N(0, τ 0, τ) L(τ 0, τ)} + 0 N(r, τ κ σ 0, τ) ε N(r, τ 0, τ) dr dr 0 [ ] N(, τ 0, τ) N(0, τ 0, τ) 0 N(r, τ {κ σ εn(, τ 0, τ) κ σ εn(0, τ 0, τ) } 0, τ) dr [ ]) = σ ε ( + κ L(τ 0, τ) {N(, τ 0, τ) N(0, τ 0, τ) } 0 N(r, τ =: σ ε{ + κ M(τ 0, τ)}. 0, τ) dr It then follows using standard theory that T / tˆφ = T û T T û T T ˆσ εt 3 T t= û t t= û t κ {N(, τ 0, τ) N(0, τ 0, τ) } κ { + κ M(τ 0, τ)}. 0 N(r, τ 0, τ) dr 9

Johan Lyhagen Department of Information Science, Uppsala University. Abstract

Johan Lyhagen Department of Information Science, Uppsala University. Abstract Why not use standard anel unit root test for testing PPP Johan Lyhagen Deartment of Information Science, Usala University Abstract In this aer we show the consequences of alying a anel unit root test that

More information

MAKING WALD TESTS WORK FOR. Juan J. Dolado CEMFI. Casado del Alisal, Madrid. and. Helmut Lutkepohl. Humboldt Universitat zu Berlin

MAKING WALD TESTS WORK FOR. Juan J. Dolado CEMFI. Casado del Alisal, Madrid. and. Helmut Lutkepohl. Humboldt Universitat zu Berlin November 3, 1994 MAKING WALD TESTS WORK FOR COINTEGRATED VAR SYSTEMS Juan J. Dolado CEMFI Casado del Alisal, 5 28014 Madrid and Helmut Lutkeohl Humboldt Universitat zu Berlin Sandauer Strasse 1 10178 Berlin,

More information

The power performance of fixed-t panel unit root tests allowing for structural breaks in their deterministic components

The power performance of fixed-t panel unit root tests allowing for structural breaks in their deterministic components ATHES UIVERSITY OF ECOOMICS AD BUSIESS DEPARTMET OF ECOOMICS WORKIG PAPER SERIES 23-203 The ower erformance of fixed-t anel unit root tests allowing for structural breaks in their deterministic comonents

More information

Hotelling s Two- Sample T 2

Hotelling s Two- Sample T 2 Chater 600 Hotelling s Two- Samle T Introduction This module calculates ower for the Hotelling s two-grou, T-squared (T) test statistic. Hotelling s T is an extension of the univariate two-samle t-test

More information

Estudios Económicos ISSN: El Colegio de México, A.C. México

Estudios Económicos ISSN: El Colegio de México, A.C. México Estudios Económicos ISSN: 0188-6916 jseme@colmex.mx El Colegio de México, A.C. México Noriega, Antonio E.; Ventosa-Santaulària, Daniel THE EFFECT OF STRUCTURAL BREAKS ON THE ENGLE-GRANGER TEST FOR COINTEGRATION

More information

A New Asymmetric Interaction Ridge (AIR) Regression Method

A New Asymmetric Interaction Ridge (AIR) Regression Method A New Asymmetric Interaction Ridge (AIR) Regression Method by Kristofer Månsson, Ghazi Shukur, and Pär Sölander The Swedish Retail Institute, HUI Research, Stockholm, Sweden. Deartment of Economics and

More information

Lower Confidence Bound for Process-Yield Index S pk with Autocorrelated Process Data

Lower Confidence Bound for Process-Yield Index S pk with Autocorrelated Process Data Quality Technology & Quantitative Management Vol. 1, No.,. 51-65, 15 QTQM IAQM 15 Lower onfidence Bound for Process-Yield Index with Autocorrelated Process Data Fu-Kwun Wang * and Yeneneh Tamirat Deartment

More information

Testing for Unit Roots in the Possible Presence of Multiple Trend Breaks Using Minimum Dickey-Fuller Statistics

Testing for Unit Roots in the Possible Presence of Multiple Trend Breaks Using Minimum Dickey-Fuller Statistics Testing for Unit Roots in the Possible Presence of Multiple Trend Breaks Using Minimum Dickey-Fuller Statistics David I. Harvey, Stephen J. Leybourne and A.M. Robert Taylor Granger Centre for Time Series

More information

Combining Logistic Regression with Kriging for Mapping the Risk of Occurrence of Unexploded Ordnance (UXO)

Combining Logistic Regression with Kriging for Mapping the Risk of Occurrence of Unexploded Ordnance (UXO) Combining Logistic Regression with Kriging for Maing the Risk of Occurrence of Unexloded Ordnance (UXO) H. Saito (), P. Goovaerts (), S. A. McKenna (2) Environmental and Water Resources Engineering, Deartment

More information

A Comparison between Biased and Unbiased Estimators in Ordinary Least Squares Regression

A Comparison between Biased and Unbiased Estimators in Ordinary Least Squares Regression Journal of Modern Alied Statistical Methods Volume Issue Article 7 --03 A Comarison between Biased and Unbiased Estimators in Ordinary Least Squares Regression Ghadban Khalaf King Khalid University, Saudi

More information

On Perron s Unit Root Tests in the Presence. of an Innovation Variance Break

On Perron s Unit Root Tests in the Presence. of an Innovation Variance Break Applied Mathematical Sciences, Vol. 3, 2009, no. 27, 1341-1360 On Perron s Unit Root ests in the Presence of an Innovation Variance Break Amit Sen Department of Economics, 3800 Victory Parkway Xavier University,

More information

Performance of lag length selection criteria in three different situations

Performance of lag length selection criteria in three different situations MPRA Munich Personal RePEc Archive Performance of lag length selection criteria in three different situations Zahid Asghar and Irum Abid Quaid-i-Azam University, Islamabad Aril 2007 Online at htts://mra.ub.uni-muenchen.de/40042/

More information

The Impact of the Initial Condition on Covariate Augmented Unit Root Tests

The Impact of the Initial Condition on Covariate Augmented Unit Root Tests The Impact of the Initial Condition on Covariate Augmented Unit Root Tests Chrystalleni Aristidou, David I. Harvey and Stephen J. Leybourne School of Economics, University of Nottingham January 2016 Abstract

More information

Estimation of the large covariance matrix with two-step monotone missing data

Estimation of the large covariance matrix with two-step monotone missing data Estimation of the large covariance matrix with two-ste monotone missing data Masashi Hyodo, Nobumichi Shutoh 2, Takashi Seo, and Tatjana Pavlenko 3 Deartment of Mathematical Information Science, Tokyo

More information

Towards understanding the Lorenz curve using the Uniform distribution. Chris J. Stephens. Newcastle City Council, Newcastle upon Tyne, UK

Towards understanding the Lorenz curve using the Uniform distribution. Chris J. Stephens. Newcastle City Council, Newcastle upon Tyne, UK Towards understanding the Lorenz curve using the Uniform distribution Chris J. Stehens Newcastle City Council, Newcastle uon Tyne, UK (For the Gini-Lorenz Conference, University of Siena, Italy, May 2005)

More information

Testing for Unit Roots in the Presence of a Possible Break in Trend and Non-Stationary Volatility

Testing for Unit Roots in the Presence of a Possible Break in Trend and Non-Stationary Volatility esting for Unit Roots in the Presence of a Possible Break in rend and Non-Stationary Volatility Giuseppe Cavaliere a, David I. Harvey b, Stephen J. Leybourne b and A.M. Robert aylor b a Department of Statistical

More information

Valid Inference in Partially Unstable GMM Models

Valid Inference in Partially Unstable GMM Models Valid Inference in Partially Unstable GMM Models Hong Li Deartment of Economics Brandeis University Waltham, MA 02454, USA hli@brandeis.edu Ulrich K. Müller Deartment of Economics Princeton University

More information

MATHEMATICAL MODELLING OF THE WIRELESS COMMUNICATION NETWORK

MATHEMATICAL MODELLING OF THE WIRELESS COMMUNICATION NETWORK Comuter Modelling and ew Technologies, 5, Vol.9, o., 3-39 Transort and Telecommunication Institute, Lomonosov, LV-9, Riga, Latvia MATHEMATICAL MODELLIG OF THE WIRELESS COMMUICATIO ETWORK M. KOPEETSK Deartment

More information

Robust methods for detecting multiple level breaks in autocorrelated time series

Robust methods for detecting multiple level breaks in autocorrelated time series Robust methods for detecting multiple level breaks in autocorrelated time series by David I. Harvey, Stephen J. Leybourne and A. M. Robert Taylor Granger Centre Discussion Paper No. 10/01 Robust Methods

More information

CHAPTER-II Control Charts for Fraction Nonconforming using m-of-m Runs Rules

CHAPTER-II Control Charts for Fraction Nonconforming using m-of-m Runs Rules CHAPTER-II Control Charts for Fraction Nonconforming using m-of-m Runs Rules. Introduction: The is widely used in industry to monitor the number of fraction nonconforming units. A nonconforming unit is

More information

Testing for Unit Roots Under Multiple Possible Trend Breaks and Non-Stationary Volatility Using Bootstrap Minimum Dickey-Fuller Statistics

Testing for Unit Roots Under Multiple Possible Trend Breaks and Non-Stationary Volatility Using Bootstrap Minimum Dickey-Fuller Statistics Testing for Unit Roots Under Multiple Possible Trend Breaks and Non-Stationary Volatility Using Bootstrap Minimum Dickey-Fuller Statistics Giuseppe Cavaliere a, David I. Harvey b, Stephen J. Leybourne

More information

Tests of the Co-integration Rank in VAR Models in the Presence of a Possible Break in Trend at an Unknown Point

Tests of the Co-integration Rank in VAR Models in the Presence of a Possible Break in Trend at an Unknown Point Tests of the Co-integration Rank in VAR Models in the Presence of a Possible Break in Trend at an Unknown Point David Harris, Steve Leybourne, Robert Taylor Monash U., U. of Nottingam, U. of Essex Economics

More information

Notes on Instrumental Variables Methods

Notes on Instrumental Variables Methods Notes on Instrumental Variables Methods Michele Pellizzari IGIER-Bocconi, IZA and frdb 1 The Instrumental Variable Estimator Instrumental variable estimation is the classical solution to the roblem of

More information

Impact Damage Detection in Composites using Nonlinear Vibro-Acoustic Wave Modulations and Cointegration Analysis

Impact Damage Detection in Composites using Nonlinear Vibro-Acoustic Wave Modulations and Cointegration Analysis 11th Euroean Conference on Non-Destructive esting (ECND 214), October 6-1, 214, Prague, Czech Reublic More Info at Oen Access Database www.ndt.net/?id=16448 Imact Damage Detection in Comosites using Nonlinear

More information

Consistent Estimation of the Number of Dynamic Factors in a Large N and T Panel

Consistent Estimation of the Number of Dynamic Factors in a Large N and T Panel Consistent Estimation of the Number of Dynamic Factors in a Large N and T Panel Dante AMENGUAL Deartment of Economics, Princeton University, Princeton, NJ 08544 (amengual@rinceton.edu) Mark W. WATSON Woodrow

More information

Estimation of Separable Representations in Psychophysical Experiments

Estimation of Separable Representations in Psychophysical Experiments Estimation of Searable Reresentations in Psychohysical Exeriments Michele Bernasconi (mbernasconi@eco.uninsubria.it) Christine Choirat (cchoirat@eco.uninsubria.it) Raffaello Seri (rseri@eco.uninsubria.it)

More information

Road Traffic Accidents in Saudi Arabia: An ARDL Approach and Multivariate Granger Causality

Road Traffic Accidents in Saudi Arabia: An ARDL Approach and Multivariate Granger Causality MPRA Munich Personal RePEc Archive Road Traffic Accidents in Saudi Arabia: An ARDL Aroach and Multivariate Granger Causality Mohammed Moosa Ageli King Saud University, RCC, Riyadh, Saudi Arabia 24. Aril

More information

Chapter 3. GMM: Selected Topics

Chapter 3. GMM: Selected Topics Chater 3. GMM: Selected oics Contents Otimal Instruments. he issue of interest..............................2 Otimal Instruments under the i:i:d: assumtion..............2. he basic result............................2.2

More information

Statics and dynamics: some elementary concepts

Statics and dynamics: some elementary concepts 1 Statics and dynamics: some elementary concets Dynamics is the study of the movement through time of variables such as heartbeat, temerature, secies oulation, voltage, roduction, emloyment, rices and

More information

On the asymptotic sizes of subset Anderson-Rubin and Lagrange multiplier tests in linear instrumental variables regression

On the asymptotic sizes of subset Anderson-Rubin and Lagrange multiplier tests in linear instrumental variables regression On the asymtotic sizes of subset Anderson-Rubin and Lagrange multilier tests in linear instrumental variables regression Patrik Guggenberger Frank Kleibergeny Sohocles Mavroeidisz Linchun Chen\ June 22

More information

Estimating Time-Series Models

Estimating Time-Series Models Estimating ime-series Models he Box-Jenkins methodology for tting a model to a scalar time series fx t g consists of ve stes:. Decide on the order of di erencing d that is needed to roduce a stationary

More information

Modeling Business Cycles with Markov Switching Arma (Ms-Arma) Model: An Application on Iranian Business Cycles

Modeling Business Cycles with Markov Switching Arma (Ms-Arma) Model: An Application on Iranian Business Cycles Modeling Business Cycles with Markov Switching Arma (Ms-Arma) Model: An Alication on Iranian Business Cycles Morteza Salehi Sarbijan 1 Faculty Member in School of Engineering, Deartment of Mechanics, Zabol

More information

System Reliability Estimation and Confidence Regions from Subsystem and Full System Tests

System Reliability Estimation and Confidence Regions from Subsystem and Full System Tests 009 American Control Conference Hyatt Regency Riverfront, St. Louis, MO, USA June 0-, 009 FrB4. System Reliability Estimation and Confidence Regions from Subsystem and Full System Tests James C. Sall Abstract

More information

On split sample and randomized confidence intervals for binomial proportions

On split sample and randomized confidence intervals for binomial proportions On slit samle and randomized confidence intervals for binomial roortions Måns Thulin Deartment of Mathematics, Usala University arxiv:1402.6536v1 [stat.me] 26 Feb 2014 Abstract Slit samle methods have

More information

Supplementary Materials for Robust Estimation of the False Discovery Rate

Supplementary Materials for Robust Estimation of the False Discovery Rate Sulementary Materials for Robust Estimation of the False Discovery Rate Stan Pounds and Cheng Cheng This sulemental contains roofs regarding theoretical roerties of the roosed method (Section S1), rovides

More information

Paper C Exact Volume Balance Versus Exact Mass Balance in Compositional Reservoir Simulation

Paper C Exact Volume Balance Versus Exact Mass Balance in Compositional Reservoir Simulation Paer C Exact Volume Balance Versus Exact Mass Balance in Comositional Reservoir Simulation Submitted to Comutational Geosciences, December 2005. Exact Volume Balance Versus Exact Mass Balance in Comositional

More information

1 Gambler s Ruin Problem

1 Gambler s Ruin Problem Coyright c 2017 by Karl Sigman 1 Gambler s Ruin Problem Let N 2 be an integer and let 1 i N 1. Consider a gambler who starts with an initial fortune of $i and then on each successive gamble either wins

More information

DEPARTMENT OF ECONOMICS ISSN DISCUSSION PAPER 20/07 TWO NEW EXPONENTIAL FAMILIES OF LORENZ CURVES

DEPARTMENT OF ECONOMICS ISSN DISCUSSION PAPER 20/07 TWO NEW EXPONENTIAL FAMILIES OF LORENZ CURVES DEPARTMENT OF ECONOMICS ISSN 1441-549 DISCUSSION PAPER /7 TWO NEW EXPONENTIAL FAMILIES OF LORENZ CURVES ZuXiang Wang * & Russell Smyth ABSTRACT We resent two new Lorenz curve families by using the basic

More information

Tests for Two Proportions in a Stratified Design (Cochran/Mantel-Haenszel Test)

Tests for Two Proportions in a Stratified Design (Cochran/Mantel-Haenszel Test) Chater 225 Tests for Two Proortions in a Stratified Design (Cochran/Mantel-Haenszel Test) Introduction In a stratified design, the subects are selected from two or more strata which are formed from imortant

More information

Steven Cook University of Wales Swansea. Abstract

Steven Cook University of Wales Swansea. Abstract On the finite sample power of modified Dickey Fuller tests: The role of the initial condition Steven Cook University of Wales Swansea Abstract The relationship between the initial condition of time series

More information

A MIXED CONTROL CHART ADAPTED TO THE TRUNCATED LIFE TEST BASED ON THE WEIBULL DISTRIBUTION

A MIXED CONTROL CHART ADAPTED TO THE TRUNCATED LIFE TEST BASED ON THE WEIBULL DISTRIBUTION O P E R A T I O N S R E S E A R C H A N D D E C I S I O N S No. 27 DOI:.5277/ord73 Nasrullah KHAN Muhammad ASLAM 2 Kyung-Jun KIM 3 Chi-Hyuck JUN 4 A MIXED CONTROL CHART ADAPTED TO THE TRUNCATED LIFE TEST

More information

Developing A Deterioration Probabilistic Model for Rail Wear

Developing A Deterioration Probabilistic Model for Rail Wear International Journal of Traffic and Transortation Engineering 2012, 1(2): 13-18 DOI: 10.5923/j.ijtte.20120102.02 Develoing A Deterioration Probabilistic Model for Rail Wear Jabbar-Ali Zakeri *, Shahrbanoo

More information

Trend and initial condition in stationarity tests: the asymptotic analysis

Trend and initial condition in stationarity tests: the asymptotic analysis Trend and initial condition in stationarity tests: the asymptotic analysis Anton Skrobotov Gaidar Institute for Economic Policy, The Russian Presidential Academy of National Economy and Public Administration

More information

Unit Roots and Structural Breaks in Panels: Does the Model Specification Matter?

Unit Roots and Structural Breaks in Panels: Does the Model Specification Matter? 18th World IMACS / MODSIM Congress, Cairns, Australia 13-17 July 2009 http://mssanz.org.au/modsim09 Unit Roots and Structural Breaks in Panels: Does the Model Specification Matter? Felix Chan 1 and Laurent

More information

Estimating function analysis for a class of Tweedie regression models

Estimating function analysis for a class of Tweedie regression models Title Estimating function analysis for a class of Tweedie regression models Author Wagner Hugo Bonat Deartamento de Estatística - DEST, Laboratório de Estatística e Geoinformação - LEG, Universidade Federal

More information

An Investigation on the Numerical Ill-conditioning of Hybrid State Estimators

An Investigation on the Numerical Ill-conditioning of Hybrid State Estimators An Investigation on the Numerical Ill-conditioning of Hybrid State Estimators S. K. Mallik, Student Member, IEEE, S. Chakrabarti, Senior Member, IEEE, S. N. Singh, Senior Member, IEEE Deartment of Electrical

More information

dn i where we have used the Gibbs equation for the Gibbs energy and the definition of chemical potential

dn i where we have used the Gibbs equation for the Gibbs energy and the definition of chemical potential Chem 467 Sulement to Lectures 33 Phase Equilibrium Chemical Potential Revisited We introduced the chemical otential as the conjugate variable to amount. Briefly reviewing, the total Gibbs energy of a system

More information

CHAPTER 5 STATISTICAL INFERENCE. 1.0 Hypothesis Testing. 2.0 Decision Errors. 3.0 How a Hypothesis is Tested. 4.0 Test for Goodness of Fit

CHAPTER 5 STATISTICAL INFERENCE. 1.0 Hypothesis Testing. 2.0 Decision Errors. 3.0 How a Hypothesis is Tested. 4.0 Test for Goodness of Fit Chater 5 Statistical Inference 69 CHAPTER 5 STATISTICAL INFERENCE.0 Hyothesis Testing.0 Decision Errors 3.0 How a Hyothesis is Tested 4.0 Test for Goodness of Fit 5.0 Inferences about Two Means It ain't

More information

Instrument endogeneity and identification-robust tests: some analytical results

Instrument endogeneity and identification-robust tests: some analytical results MPRA Munich Personal RePEc Archive Instrument endogeneity and identification-robust tests: some analytical results Firmin Sabro Doko chatoka and Jean-Marie Dufour 3. May 2008 Online at htt://mra.ub.uni-muenchen.de/2963/

More information

A Simple Panel Stationarity Test in the Presence of Cross-Sectional Dependence

A Simple Panel Stationarity Test in the Presence of Cross-Sectional Dependence A Simle Panel Stationarity est in the Presence of Cross-Sectional Deendence Kaddour Hadri Eiji Kurozumi 2 Queen s University Management School Deartment of Economics Queen s University Hitotsubashi University

More information

Partial Identification in Triangular Systems of Equations with Binary Dependent Variables

Partial Identification in Triangular Systems of Equations with Binary Dependent Variables Partial Identification in Triangular Systems of Equations with Binary Deendent Variables Azeem M. Shaikh Deartment of Economics University of Chicago amshaikh@uchicago.edu Edward J. Vytlacil Deartment

More information

Using the Divergence Information Criterion for the Determination of the Order of an Autoregressive Process

Using the Divergence Information Criterion for the Determination of the Order of an Autoregressive Process Using the Divergence Information Criterion for the Determination of the Order of an Autoregressive Process P. Mantalos a1, K. Mattheou b, A. Karagrigoriou b a.deartment of Statistics University of Lund

More information

On Wald-Type Optimal Stopping for Brownian Motion

On Wald-Type Optimal Stopping for Brownian Motion J Al Probab Vol 34, No 1, 1997, (66-73) Prerint Ser No 1, 1994, Math Inst Aarhus On Wald-Tye Otimal Stoing for Brownian Motion S RAVRSN and PSKIR The solution is resented to all otimal stoing roblems of

More information

Questions and Answers on Unit Roots, Cointegration, VARs and VECMs

Questions and Answers on Unit Roots, Cointegration, VARs and VECMs Questions and Answers on Unit Roots, Cointegration, VARs and VECMs L. Magee Winter, 2012 1. Let ɛ t, t = 1,..., T be a series of independent draws from a N[0,1] distribution. Let w t, t = 1,..., T, be

More information

Finite Mixture EFA in Mplus

Finite Mixture EFA in Mplus Finite Mixture EFA in Mlus November 16, 2007 In this document we describe the Mixture EFA model estimated in Mlus. Four tyes of deendent variables are ossible in this model: normally distributed, ordered

More information

Uniform Law on the Unit Sphere of a Banach Space

Uniform Law on the Unit Sphere of a Banach Space Uniform Law on the Unit Shere of a Banach Sace by Bernard Beauzamy Société de Calcul Mathématique SA Faubourg Saint Honoré 75008 Paris France Setember 008 Abstract We investigate the construction of a

More information

Location of solutions for quasi-linear elliptic equations with general gradient dependence

Location of solutions for quasi-linear elliptic equations with general gradient dependence Electronic Journal of Qualitative Theory of Differential Equations 217, No. 87, 1 1; htts://doi.org/1.14232/ejqtde.217.1.87 www.math.u-szeged.hu/ejqtde/ Location of solutions for quasi-linear ellitic equations

More information

Shadow Computing: An Energy-Aware Fault Tolerant Computing Model

Shadow Computing: An Energy-Aware Fault Tolerant Computing Model Shadow Comuting: An Energy-Aware Fault Tolerant Comuting Model Bryan Mills, Taieb Znati, Rami Melhem Deartment of Comuter Science University of Pittsburgh (bmills, znati, melhem)@cs.itt.edu Index Terms

More information

Lecture 2: Consistency of M-estimators

Lecture 2: Consistency of M-estimators Lecture 2: Instructor: Deartment of Economics Stanford University Preared by Wenbo Zhou, Renmin University References Takeshi Amemiya, 1985, Advanced Econometrics, Harvard University Press Newey and McFadden,

More information

Heteroskedasticity, Autocorrelation, and Spatial Correlation Robust Inference in Linear Panel Models with Fixed-E ects

Heteroskedasticity, Autocorrelation, and Spatial Correlation Robust Inference in Linear Panel Models with Fixed-E ects Heteroskedasticity, Autocorrelation, and Satial Correlation Robust Inference in Linear Panel Models with Fixed-E ects Timothy J. Vogelsang Deartments of Economics, Michigan State University December 28,

More information

Lecture 3 Consistency of Extremum Estimators 1

Lecture 3 Consistency of Extremum Estimators 1 Lecture 3 Consistency of Extremum Estimators 1 This lecture shows how one can obtain consistency of extremum estimators. It also shows how one can find the robability limit of extremum estimators in cases

More information

Asymptotic F Test in a GMM Framework with Cross Sectional Dependence

Asymptotic F Test in a GMM Framework with Cross Sectional Dependence Asymtotic F Test in a GMM Framework with Cross Sectional Deendence Yixiao Sun Deartment of Economics University of California, San Diego Min Seong Kim y Deartment of Economics Ryerson University First

More information

Linear diophantine equations for discrete tomography

Linear diophantine equations for discrete tomography Journal of X-Ray Science and Technology 10 001 59 66 59 IOS Press Linear diohantine euations for discrete tomograhy Yangbo Ye a,gewang b and Jiehua Zhu a a Deartment of Mathematics, The University of Iowa,

More information

Radial Basis Function Networks: Algorithms

Radial Basis Function Networks: Algorithms Radial Basis Function Networks: Algorithms Introduction to Neural Networks : Lecture 13 John A. Bullinaria, 2004 1. The RBF Maing 2. The RBF Network Architecture 3. Comutational Power of RBF Networks 4.

More information

Darmstadt Discussion Papers in Economics

Darmstadt Discussion Papers in Economics Darmstadt Discussion Papers in Economics The Effect of Linear Time Trends on Cointegration Testing in Single Equations Uwe Hassler Nr. 111 Arbeitspapiere des Instituts für Volkswirtschaftslehre Technische

More information

A TIME SERIES PARADOX: UNIT ROOT TESTS PERFORM POORLY WHEN DATA ARE COINTEGRATED

A TIME SERIES PARADOX: UNIT ROOT TESTS PERFORM POORLY WHEN DATA ARE COINTEGRATED A TIME SERIES PARADOX: UNIT ROOT TESTS PERFORM POORLY WHEN DATA ARE COINTEGRATED by W. Robert Reed Department of Economics and Finance University of Canterbury, New Zealand Email: bob.reed@canterbury.ac.nz

More information

Asymptotically Optimal Simulation Allocation under Dependent Sampling

Asymptotically Optimal Simulation Allocation under Dependent Sampling Asymtotically Otimal Simulation Allocation under Deendent Samling Xiaoing Xiong The Robert H. Smith School of Business, University of Maryland, College Park, MD 20742-1815, USA, xiaoingx@yahoo.com Sandee

More information

Unobservable Selection and Coefficient Stability: Theory and Evidence

Unobservable Selection and Coefficient Stability: Theory and Evidence Unobservable Selection and Coefficient Stability: Theory and Evidence Emily Oster Brown University and NBER August 9, 016 Abstract A common aroach to evaluating robustness to omitted variable bias is to

More information

arxiv: v1 [physics.data-an] 26 Oct 2012

arxiv: v1 [physics.data-an] 26 Oct 2012 Constraints on Yield Parameters in Extended Maximum Likelihood Fits Till Moritz Karbach a, Maximilian Schlu b a TU Dortmund, Germany, moritz.karbach@cern.ch b TU Dortmund, Germany, maximilian.schlu@cern.ch

More information

A Time-Varying Threshold STAR Model of Unemployment

A Time-Varying Threshold STAR Model of Unemployment A Time-Varying Threshold STAR Model of Unemloyment michael dueker a michael owyang b martin sola c,d a Russell Investments b Federal Reserve Bank of St. Louis c Deartamento de Economia, Universidad Torcuato

More information

Econ 3790: Business and Economics Statistics. Instructor: Yogesh Uppal

Econ 3790: Business and Economics Statistics. Instructor: Yogesh Uppal Econ 379: Business and Economics Statistics Instructor: Yogesh Ual Email: yual@ysu.edu Chater 9, Part A: Hyothesis Tests Develoing Null and Alternative Hyotheses Tye I and Tye II Errors Poulation Mean:

More information

Adaptive estimation with change detection for streaming data

Adaptive estimation with change detection for streaming data Adative estimation with change detection for streaming data A thesis resented for the degree of Doctor of Philosohy of the University of London and the Diloma of Imerial College by Dean Adam Bodenham Deartment

More information

Economic modelling and forecasting. 2-6 February 2015

Economic modelling and forecasting. 2-6 February 2015 Economic modelling and forecasting 2-6 February 2015 Bank of England 2015 Ole Rummel Adviser, CCBS at the Bank of England ole.rummel@bankofengland.co.uk Philosophy of my presentations Everything should

More information

A Model for Randomly Correlated Deposition

A Model for Randomly Correlated Deposition A Model for Randomly Correlated Deosition B. Karadjov and A. Proykova Faculty of Physics, University of Sofia, 5 J. Bourchier Blvd. Sofia-116, Bulgaria ana@hys.uni-sofia.bg Abstract: A simle, discrete,

More information

Use of Transformations and the Repeated Statement in PROC GLM in SAS Ed Stanek

Use of Transformations and the Repeated Statement in PROC GLM in SAS Ed Stanek Use of Transformations and the Reeated Statement in PROC GLM in SAS Ed Stanek Introduction We describe how the Reeated Statement in PROC GLM in SAS transforms the data to rovide tests of hyotheses of interest.

More information

Causality Testing using Higher Order Statistics

Causality Testing using Higher Order Statistics Causality Testing using Higher Order Statistics Dr Sanya Dudukovic International Management Deartment Franklin College, Switzerland Fax: 41 91 994 41 17 E-mail : Sdudukov@fc.edu Abstract : A new causality

More information

Journal of Chemical and Pharmaceutical Research, 2014, 6(5): Research Article

Journal of Chemical and Pharmaceutical Research, 2014, 6(5): Research Article Available online www.jocr.com Journal of Chemical and Pharmaceutical Research, 204, 6(5):580-585 Research Article ISSN : 0975-7384 CODEN(USA) : JCPRC5 Exort facilitation and comarative advantages of the

More information

State Estimation with ARMarkov Models

State Estimation with ARMarkov Models Deartment of Mechanical and Aerosace Engineering Technical Reort No. 3046, October 1998. Princeton University, Princeton, NJ. State Estimation with ARMarkov Models Ryoung K. Lim 1 Columbia University,

More information

AI*IA 2003 Fusion of Multiple Pattern Classifiers PART III

AI*IA 2003 Fusion of Multiple Pattern Classifiers PART III AI*IA 23 Fusion of Multile Pattern Classifiers PART III AI*IA 23 Tutorial on Fusion of Multile Pattern Classifiers by F. Roli 49 Methods for fusing multile classifiers Methods for fusing multile classifiers

More information

Monopolist s mark-up and the elasticity of substitution

Monopolist s mark-up and the elasticity of substitution Croatian Oerational Research Review 377 CRORR 8(7), 377 39 Monoolist s mark-u and the elasticity of substitution Ilko Vrankić, Mira Kran, and Tomislav Herceg Deartment of Economic Theory, Faculty of Economics

More information

Feedback-error control

Feedback-error control Chater 4 Feedback-error control 4.1 Introduction This chater exlains the feedback-error (FBE) control scheme originally described by Kawato [, 87, 8]. FBE is a widely used neural network based controller

More information

A Note on Massless Quantum Free Scalar Fields. with Negative Energy Density

A Note on Massless Quantum Free Scalar Fields. with Negative Energy Density Adv. Studies Theor. Phys., Vol. 7, 13, no. 1, 549 554 HIKARI Ltd, www.m-hikari.com A Note on Massless Quantum Free Scalar Fields with Negative Energy Density M. A. Grado-Caffaro and M. Grado-Caffaro Scientific

More information

arxiv: v2 [stat.me] 3 Nov 2014

arxiv: v2 [stat.me] 3 Nov 2014 onarametric Stein-tye Shrinkage Covariance Matrix Estimators in High-Dimensional Settings Anestis Touloumis Cancer Research UK Cambridge Institute University of Cambridge Cambridge CB2 0RE, U.K. Anestis.Touloumis@cruk.cam.ac.uk

More information

Asymptotic Properties of the Markov Chain Model method of finding Markov chains Generators of..

Asymptotic Properties of the Markov Chain Model method of finding Markov chains Generators of.. IOSR Journal of Mathematics (IOSR-JM) e-issn: 78-578, -ISSN: 319-765X. Volume 1, Issue 4 Ver. III (Jul. - Aug.016), PP 53-60 www.iosrournals.org Asymtotic Proerties of the Markov Chain Model method of

More information

Characterizing the Behavior of a Probabilistic CMOS Switch Through Analytical Models and Its Verification Through Simulations

Characterizing the Behavior of a Probabilistic CMOS Switch Through Analytical Models and Its Verification Through Simulations Characterizing the Behavior of a Probabilistic CMOS Switch Through Analytical Models and Its Verification Through Simulations PINAR KORKMAZ, BILGE E. S. AKGUL and KRISHNA V. PALEM Georgia Institute of

More information

Unit Roots in Time Series with Changepoints

Unit Roots in Time Series with Changepoints International Journal of Statistics and Probability; Vol. 6, No. 6; November 2017 ISSN 1927-7032 E-ISSN 1927-7040 Published by Canadian Center of Science and Education Unit Roots in Time Series with Changepoints

More information

4. Score normalization technical details We now discuss the technical details of the score normalization method.

4. Score normalization technical details We now discuss the technical details of the score normalization method. SMT SCORING SYSTEM This document describes the scoring system for the Stanford Math Tournament We begin by giving an overview of the changes to scoring and a non-technical descrition of the scoring rules

More information

Moreover, the second term is derived from: 1 T ) 2 1

Moreover, the second term is derived from: 1 T ) 2 1 170 Moreover, the second term is derived from: 1 T T ɛt 2 σ 2 ɛ. Therefore, 1 σ 2 ɛt T y t 1 ɛ t = 1 2 ( yt σ T ) 2 1 2σ 2 ɛ 1 T T ɛt 2 1 2 (χ2 (1) 1). (b) Next, consider y 2 t 1. T E y 2 t 1 T T = E(y

More information

NONLINEAR OPTIMIZATION WITH CONVEX CONSTRAINTS. The Goldstein-Levitin-Polyak algorithm

NONLINEAR OPTIMIZATION WITH CONVEX CONSTRAINTS. The Goldstein-Levitin-Polyak algorithm - (23) NLP - NONLINEAR OPTIMIZATION WITH CONVEX CONSTRAINTS The Goldstein-Levitin-Polya algorithm We consider an algorithm for solving the otimization roblem under convex constraints. Although the convexity

More information

Testing for non-stationarity

Testing for non-stationarity 20 November, 2009 Overview The tests for investigating the non-stationary of a time series falls into four types: 1 Check the null that there is a unit root against stationarity. Within these, there are

More information

A Simle Panel Stationarity itle Cross-Sectional Deendence est in Author(s) Hadri, Kaddour; Kurozumi, Eiji Citation Issue 00-06 Date ye echnical Reort ext Version ublisher URL htt://hdl.handle.net/0086/8605

More information

An Improved Generalized Estimation Procedure of Current Population Mean in Two-Occasion Successive Sampling

An Improved Generalized Estimation Procedure of Current Population Mean in Two-Occasion Successive Sampling Journal of Modern Alied Statistical Methods Volume 15 Issue Article 14 11-1-016 An Imroved Generalized Estimation Procedure of Current Poulation Mean in Two-Occasion Successive Samling G. N. Singh Indian

More information

Nonsense Regressions due to Neglected Time-varying Means

Nonsense Regressions due to Neglected Time-varying Means Nonsense Regressions due to Neglected Time-varying Means Uwe Hassler Free University of Berlin Institute of Statistics and Econometrics Boltzmannstr. 20 D-14195 Berlin Germany email: uwe@wiwiss.fu-berlin.de

More information

MODELING THE RELIABILITY OF C4ISR SYSTEMS HARDWARE/SOFTWARE COMPONENTS USING AN IMPROVED MARKOV MODEL

MODELING THE RELIABILITY OF C4ISR SYSTEMS HARDWARE/SOFTWARE COMPONENTS USING AN IMPROVED MARKOV MODEL Technical Sciences and Alied Mathematics MODELING THE RELIABILITY OF CISR SYSTEMS HARDWARE/SOFTWARE COMPONENTS USING AN IMPROVED MARKOV MODEL Cezar VASILESCU Regional Deartment of Defense Resources Management

More information

Long memory and changing persistence

Long memory and changing persistence Long memory and changing persistence Robinson Kruse and Philipp Sibbertsen August 010 Abstract We study the empirical behaviour of semi-parametric log-periodogram estimation for long memory models when

More information

SHU-PING SHI. School of Economics, The Australian National University SUMMARY

SHU-PING SHI. School of Economics, The Australian National University SUMMARY ESING FOR PERIODICALLY COLLAPSING BUBBLES: AN GENERALIZED SUP ADF ES SHU-PING SHI School of Economics, he Australian National University SUMMARY Identifying exlosive bubbles under the influence of their

More information

Robust and powerful tests for nonlinear deterministic components

Robust and powerful tests for nonlinear deterministic components Robust and powerful tests for nonlinear deterministic components Sam Astill a, David I. Harvey b, Stephen J. Leybourne b and A.M. Robert Taylor c a. Department of Economics, University of Warwick. b. Granger

More information

TESTING FOR UNIT ROOTS IN PANELS IN THE PRESENCE OF STRUCTURAL CHANGE WITH AN APPLICATION TO OECD UNEMPLOYMENT

TESTING FOR UNIT ROOTS IN PANELS IN THE PRESENCE OF STRUCTURAL CHANGE WITH AN APPLICATION TO OECD UNEMPLOYMENT ESING FOR UNI ROOS IN PANELS IN HE PRESENCE OF SRUCURAL CHANGE WIH AN APPLICAION O OECD UNEMPLOYMEN Christian J. Murray a and David H. Papell b a Department of Economics, University of Houston, Houston,

More information

Hidden Predictors: A Factor Analysis Primer

Hidden Predictors: A Factor Analysis Primer Hidden Predictors: A Factor Analysis Primer Ryan C Sanchez Western Washington University Factor Analysis is a owerful statistical method in the modern research sychologist s toolbag When used roerly, factor

More information

A Qualitative Event-based Approach to Multiple Fault Diagnosis in Continuous Systems using Structural Model Decomposition

A Qualitative Event-based Approach to Multiple Fault Diagnosis in Continuous Systems using Structural Model Decomposition A Qualitative Event-based Aroach to Multile Fault Diagnosis in Continuous Systems using Structural Model Decomosition Matthew J. Daigle a,,, Anibal Bregon b,, Xenofon Koutsoukos c, Gautam Biswas c, Belarmino

More information