A PRIMER ON REGRESSION SPLINES. 1. Overview
|
|
- Neal Ramsey
- 8 years ago
- Views:
Transcription
1 A PRIMER ON REGRESSION SPLINES JEFFREY S. RACINE 1. Overvew B-splnes consttute an appealng method for the nonparametrc estmaton of a range of statstcal objects of nterest. In ths prmer we focus our attenton on the estmaton of a condtonal mean,.e. the regresson functon. A splne s a functon that s constructed pece-wse from polynomal functons. The term comes from the tool used by shpbulders and drafters to construct smooth shapes havng desred propertes. Drafters have long made use of a bendable strp fxed n poston at a number of ponts that relaxes to form a smooth curve passng through those ponts. The malleablty of the splne materal combned wth the constrant of the control ponts would cause the strp to take the shape that mnmzed the energy requred for bendng t between the fxed ponts, ths beng the smoothest possble shape. We shall rely on a class of splnes called B-splnes ( bass-splnes ). A B-splne functon s the maxmally dfferentable nterpolatve bass functon. The B-splne s a generalzaton of the Bézer curve (a B-splne wth no nteror knots s a Bézer curve). B-splnes are defned by ther order m and number of nteror knots N (there are two endponts whch are themselves knots so the total number of knots wll be N+2). The degree of the B-splne polynomal wll be the splne order m mnus one (degree = m 1). To best apprecate the nature of B-splnes, we shall frst consder a smple type of splne, the Bézer functon, and then move on to the more flexble and powerful generalzaton, the B-splne tself. We begn wth the unvarate case n Secton 2 where we consder the unvarate Bézer functon. In Secton 3 we turn to the unvarate B-splne functon, and then n Secton 4 we turn to the multvarate case where we also brefly menton how one could handle the presence of categorcal predctors. We presume that nterest les n regresson splne methodology whch dffers n a number of ways from smoothng splnes, both of whch are popular n appled settngs. The fundamental dfference between the two approaches s that smoothng splnes explctly penalze roughness and use the data ponts themselves as potental knots whereas regresson splnes place knots at equdstant/eququantle ponts. We drect the nterested reader to Wahba (1990) for a treatment of smoothng splnes. Date: December 18, These notes are culled from a varety of sources. I am solely responsble for all errors. Suggestons are welcomed (racnej@mcmaster.ca). 1
2 2 JEFFREY S. RACINE 2. Bézer curves We present an overvew of Bézer curves whch form the bass for the B-splnes that follow. We begn wth a smple llustraton, that of a quadratc Bézer curve. Example 2.1. A quadratc Bézer curve. A quadratc Bézer curve s the path traced by the functon B(x), gven ponts β 0, β 1, and β 2, where B(x) = β 0 (1 x) 2 +2β 1 (1 x)x+β 2 x 2 = 2 β B (x), x [0,1]. The terms B 0 (x) = (1 x) 2, B 1 (x) = 2(1 x)x, and B 2 (x) = x 2 are the bases whch s ths case turn out to be Bernsten polynomals (Bernsten (1912)). For our purposes the control ponts β, = 0,1,2, wll be parameters that could be selected by least squares fttng n a regresson settng, but more on that later. Consder the followng smple example where we plot a quadratc Bézer curve wth arbtrary control ponts: B(x) For ths smple llustraton we set β 0 = 1, β 1 = 1, β 2 = 2. Note that the dervatve of ths curve s B (x) = 2(1 x)(β 1 β 0 )+2x(β 2 β 1 ), whch s a polynomal of degree one. Ths example of a Bézer curve wll also be seen to be a second-degree B-splne wth no nteror knots or, equvalently, a thrd-order B-splne wth no nteror knots. Usng the termnology of B-splnes, n ths example we have a thrd-order B-splne (m = 3) whch s of polynomal degree two (m 1 = 2) havng hghest dervatve of polynomal degree one (m 2 = 1). x
3 A PRIMER ON REGRESSION SPLINES The Bézer curve defned. More generally, a Bézer curve of degree n(order m) s composed of m = n+1 terms and s gven by n ( ) n B(x) = β (1 x) n x (1) = n β B,n (x), where ( ) n = n! (n )!!, whch can be expressed recursvely as ( n 1 ) ( n ) B(x) = (1 x) β B,n 1 (x) +x β B,n 1 (x), so a degree n Bézer curve s a lnear nterpolaton between two degree n 1 Bézer curves. =1 Example 2.2. A quadratc Bézer curve as a lnear nterpolaton between two lnear Bézer curves. The lnear Bézer curve s gven by β 0 (1 x)+β 1 x, and above we showed that the quadratc Bézer curve s gven by β 0 (1 x) 2 +2β 1 (1 x)x+β 2 x 2. So, when n = 2 (quadratc), we have B(x) = (1 x)(β 0 (1 x)+β 1 x)+x(β 1 (1 x)+β 2 x) = β 0 (1 x) 2 +2β 1 (1 x)x+β 2 x 2. Ths s essentally a modfed verson of the dea of takng lnear nterpolatons of lnear nterpolatons of lnear nterpolatons and so on. Note that the polynomals ( ) n B,n (x) = (1 x) n x are called Bernsten bass polynomals of degree n and are such that n B,n(x) = 1, unlke raw polynomals. 1 The m = n + 1 control ponts β, = 0,...,n, are somewhat ancllary to the dscusson here, but wll fgure promnently when we turn to regresson as n a regresson settng they wll be the coeffcents of the regresson model. 1 Naturally we defne x 0 = (1 x) 0 = 1, and by raw polynomals we smply mean x j, j = 0,...,n.
4 4 JEFFREY S. RACINE Example 2.3. The quadratc Bézer curve bass functons. The fgure below presents the bases B,n (x) underlyng a Bézer curve for = 1,...,2 and n = 2. B(x) x These bases are B 0,2 (x) = (1 x) 2, B 1,2 (x) = 2(1 x)x, and B 2,2 (x) = x 2 and llustrate the foundaton upon whch the Bézer curves are bult Dervatves of splne functons. From de Boor (2001) we know that the dervatves of splne functons can be smply expressed n terms of lower order splne functons. In partcular, for the Bézer curve we have where β (0) n l B (l) (x) = β (l) B,n l (x), = β, 0 n, and ( ) β (l) = (n l) β (l 1) +1 β (l 1) /(t t n+l ), 0 n l. See Zhou & Wolfe (2000) for detals. We now turn our attenton to the B-splne functon. Ths can be thought of as a generalzaton of the Bézer curve where we now allow for there to be addtonal breakponts called nteror knots. 3. B-splnes 3.1. B-splne knots. B-splne curves are composed from many polynomal peces and are therefore more versatle than Bézer curves. Consder N +2 real values t, called knots (N 0 are called nteror knots and there are always two endponts, t 0 and t N+1 ), wth t 0 t 1 t N+1. When the knots are equdstant they are sad to be unform, otherwse they are sad to be nonunform. One popular type of knot s the quantle knot sequence where the nteror knots are the quantles from the emprcal dstrbuton of the underlyng varable. Quantle knots guarantee that
5 A PRIMER ON REGRESSION SPLINES 5 an equal number of sample observatons le n each nterval whle the ntervals wll have dfferent lengths (as opposed to dfferent numbers of ponts lyng n equal length ntervals). Bézer curves possess two endpont knots, t 0 and t 1, and no nteror knots hence are a lmtng case,.e. a B-splne for whch N = The B-splne bass functon. Let t = {t Z be a sequence of non-decreasng real numbers (t t +1 ) such that 2 Defne the augmented the knot set t 0 t 1 t N+1. t (m 1) = = t 0 t 1 t N t N+1 = = t N+m, where we have appended the lower and upper boundary knots t 0 and t 1 n = m 1 tmes (ths s needed due to the recursve nature of the B-splne). If we wanted we could then reset the ndex for the frst element of the augmented knot set (.e. t (m 1) ) so that the N +2m augmented knots t are now ndexed by = 0,...,N +2m 1 (see the example below for an llustraton). Foreachoftheaugmentedknotst, = 0,...,N+2m 1, werecursvelydefneasetofreal-valued functons B,j (for j = 0,1,...,n, n beng the degree of the B-splne bass) as follows: B,0 (x) = { 1 f t x < t +1 0 otherwse. where B,j+1 (x) = α,j+1 (x)b,j (x)+[1 α +1,j+1 (x)]b +1,j (x), For the above computaton we defne 0/0 as 0. Defntons. Usng the notaton above: x t f t +j t α,j (x) = t +j t 0 otherwse. (1) the sequence t s known as a knot sequence, and the ndvdual term n the sequence s a knot. (2) the functons B,j are called the -th B-splne bass functons of order j, and the recurrence relaton s called the de Boor recurrence relaton, after ts dscoverer Carl de Boor (de Boor (2001)). (3) gven any non-negatve nteger j, the vector space V j (t) over R, generated by the set of all B-splne bass functons of order j s called the B-splne of order j. In other words, the B-splne V j (t) = span{b,j (x) = 0,1,... over R. (4) Any element of V j (t) s a B-splne functon of order j. 2 Ths descrpton s based upon the dscusson found at
6 6 JEFFREY S. RACINE The frst term B 0,n s often referred to as the ntercept. In typcal splne mplementatons the opton ntercept=false denotes droppng ths term whle ntercept=true denotes keepng t (recall that n B,n(x) = 1 whch can lead to perfect multcollnearty n a regresson settng; also see Zhou & Wolfe (2000) who nstead apply shrnkage methods). Example 3.4. A fourth-order B-splne bass functon wth three nteror knots and ts frst dervatve functon. Suppose there are N = 3 nteror knots gven by (0.25,0.5,0.75), the boundary knots are (0,1), and the degree of the splne s n = 3 hence the order s m = 4. The set of all knot ponts needed to construct the B-splne s (0,0,0,0,0.25,0.5,0.75,1,1,1,1) and the number of bass functons s K = N +m = 7. The seven cubc splne bass functons wll be denoted B 0,3,...,B 6,3. The fgure below presents ths example of a thrd degree B-splne wth three nteror knots along wth ts frst dervatve (the splne dervatves would be requred n order to compute dervatves from the splne regresson model). B B.derv x x To summarze, n ths llustraton we have an order m = 4 (degree = 3) B-splne (left) wth 4 sub-ntervals (segments) usng unform knots (N = 3 nteror knots, 5 knots n total (2 endpont knots)) and ts 1st-order dervatve (rght). The dmenson of B(x) s K = N +m = 7. See the appendx for R code (R Development Core Team (2011)) that mplements the B-splne bass functon The B-splne functon. A B-splne of degree n (of splne order m = n+1) s a parametrc curve composed of a lnear combnaton of bass B-splnes B,n (x) of degree n gven by (2) B(x) = N+n β B,n (x), x [t 0,t N+1 ].
7 A PRIMER ON REGRESSION SPLINES 7 The β are called control ponts or de Boor ponts. For an order m B-splne havng N nteror knots there are K = N +m = N +n+1 control ponts (one when j = 0). The B-splne order m must be at least 2 (hence at least lnear,.e. degree n s at least 1) and the number of nteror knots must be non-negatve (N 0). See the appendx for R code (R Development Core Team (2011)) that mplements the B-splne functon. 4. Multvarate B-splne regresson The functonal form of parametrc regresson models must naturally be specfed by the user. Typcally practtoners rely on raw polynomals and also often choose the form of the regresson functon (.e. the order of the polynomal for each predctor) n an ad-hoc manner. However, raw polynomals are not suffcently flexble for our purposes, partcularly because they possess no nteror knots whch s where B-splnes derve ther unque propertes. Furthermore, n a regresson settng we typcally encounter multple predctors whch can be contnuous or categorcal n nature, and tradtonal splnes are for contnuous predctors. Below we brefly descrbe a multvarate kernel weghted tensor product B-splne regresson method(kernel weghtng s used to handle the presence of the categorcal predctors). Ths method s mplemented n the R package crs (Racne & Ne (2011)) Multvarate knots, ntervals, and splne bases. In general we wll have q predctors, X = (X 1,...,X q ) T. We assume that each X l, 1 l q, s dstrbuted on a compact nterval [a l,b l ], and wthout loss of generalty, we take all ntervals [a l,b l ] = [0,1]. Let G l = G (m l 2) l be the space of polynomal splnes of order m l. We note that G l conssts of functons satsfyng () s a polynomal of degree m l 1 on each of the sub-ntervals I jl,l,j l = 0,...,N l ; () for m l 2, s m l 2 tmes contnuously dfferentable on [0,1]. Pre-select an nteger N l = N n,l. Dvde [a l,b l ] = [0,1] nto (N l +1) sub-ntervals I jl,l = [t jl,l,t jl +1,l), j l = 0,...,N l 1, I Nl,l = [t Nl,l,1], where {t jl,l N l j l =1 s a sequence of equally-spaced ponts, called nteror knots, gven as t (ml 1),l = = t 0,l = 0 < t 1,l < < t Nl,l < 1 = t Nl +1,l = = t Nl +m l,l, n whch t jl,l = j l h l, j l = 0,1...,N l +1, h l = 1/(N l +1) s the dstance between neghborng knots. Let K l = K n,l = N l +m l, where N l s the number of nteror knots and m l s the splne order, and let B l (x l ) = {B jl,l(x l ) : 1 m l j l N l T be a bass system of the space G l.
8 8 JEFFREY S. RACINE We defne the space of tensor-product polynomal splnes by G = q l=1 G l. It s clear that G s a lnear space of dmenson K n = q l=1 K l. Then 3 [ {Bj1 B(x) =,...,j q (x) N 1,...,N q j 1 =1 m 1,...,j q=1 m q ]K = B 1(x 1 ) B q (x q ) n 1 s a bass system of the space G, where x =(x l ) q l=1. Let B = [{B(X 1 ),...,B(X n ) T] n K n Splne regresson. In what follows we presume that the reader s nterested n the unknown condtonal mean n the followng locaton-scale model, (3) Y = g(x,z)+σ(x,z)ε, where g( ) s an unknown functon, X =(X 1,...,X q ) T s a q-dmensonal vector of contnuous predctors, and Z = (Z 1,...,Z r ) T s an r-dmensonal vector of categorcal predctors. Lettng z = (z s ) r s=1, we assume that z s takes c s dfferent values n D s {0,1,...,c s 1, s = 1,...,r, and let c s be a fnte postve constant. Let ( Y,X T ) n,zt =1 be an..d copy of ( Y,X T,Z T). Assume for 1 l q, each X l s dstrbuted on a compact nterval [a l,b l ], and wthout loss of generalty, we take all ntervals [a l,b l ] = [0,1]. In order to handle the presence of categorcal predctors, we defne (4) { 1,when Z s = z s l(z s,z s,λ s ) =, λ s, otherwse. r r L(Z,z,λ) = l(z s,z s,λ s ) = s=1 s=1 λ 1(Zs zs) s, where l( ) s a varant of Atchson & Atken s (1976) unvarate categorcal kernel functon, L( ) s a product categorcal kernel functon, and λ = (λ 1,λ 2,...,λ r ) T s the vector of bandwdths for each of the categorcal predctors. See Ma, Racne & Yang (under revson) and Ma & Racne (2013) for further detals. We estmate β(z) by mnmzng the followng weghted least squares crteron, β(z) = arg mn β R Kn n =1 { Y B(X ) T β 2L(Z,z,λ). Let L z = dag{l(z 1,z,λ),...,L(Z n,z,λ) be a dagonal matrx wth L(Z,z,λ), 1 n as the dagonal entres. Then β(z) can be wrtten as (5) β(z) = ( n 1 B T L z B ) 1( n 1 B T L z Y ), 3 The notaton here may throw off those used to sums of the form n =1, n > 0 (.e. sum ndces that are postve ntegers), so consder a smple llustraton that may defuse ths ssue. Suppose there are no nteror knots (N = 0) and we consder a quadratc (degree n equal to two hence the splne order s three). Then N =1 m contans three terms havng ndces = 2, 1,0. In general the number of terms s the number the number of nteror knots N plus the splne order m, whch we denote K = N +m. We could alternatvely sum from 1 to N +m, or from 0 to N + m 1 of from 0 to N + n (the latter beng consstent wth the Bézer curve defnton n (1) and the B-splne defnton n (2)).
9 A PRIMER ON REGRESSION SPLINES 9 where Y =(Y 1,...,Y n ) T. g(x,z) s estmated by ĝ(x,z) = B(x) T β(z). See the appendx for R code (R Development Core Team (2011)) that mplements the B-splne bass functon and then uses least squares to construct the regresson model for a smulated data generatng process. References Atchson, J. & Atken, C. G. G. (1976), Multvarate bnary dscrmnaton by the kernel method, Bometrka 63(3), Bernsten, S. (1912), Démonstraton du théorème de Weerstrass fonde sur le calcul des probabltes, Comm. Soc. Math. Kharkov 13, 1 2. de Boor, C. (2001), A practcal gude to splnes, Sprnger. Ma, S. & Racne, J. S. (2013), Addtve regresson splnes wth rrelevant categorcal and contnuous regressors, Statstca Snca 23, Ma, S., Racne, J. S. & Yang, L. (under revson), Splne regresson n the presence of categorcal predctors, Journal of Appled Econometrcs. R Development Core Team (2011), R: A Language and Envronment for Statstcal Computng, R Foundaton for Statstcal Computng, Venna, Austra. ISBN URL: Racne, J. S. & Ne, Z. (2011), crs: Categorcal Regresson Splnes. R package verson Wahba, G. (1990), Splne Models for Observatonal Data, SIAM. Zhou, S. & Wolfe, D. A. (2000), On dervatve estmaton n splne regresson, Statstca Snca 10,
10 10 JEFFREY S. RACINE Appendx A. Sample R code for constructng B-splnes The followng code uses recurson to compute the B-splne bass and B-splne functon. Then a smple llustraton demonstrates how one could mmedately compute a least-squares ft usng the B-splne. In the sprt of recurson, t has been sad that To terate s human; to recurse dvne. (L. Peter Deutsch). R Code for Implementng B-splne bass functons and the B-splne tself. ## $Id: splne_prmer.rnw,v /01/22 17:43:52 jracne Exp jracne $ ## Aprl The code below s based upon an llustraton that ## can be found n ## by Dr. Samran Snha (Department of Statstcs, Texas A&M). I am ## solely to blame for any errors and can be contacted at ## racnej@mcmaster.ca (Jeffrey S. Racne). ## Ths functon s a (smplfed) R mplementaton of the bs() ## functon n the splnes lbrary and llustrates how the Cox-de Boor ## recurson formula s used to construct B-splnes. bass <- functon(x, degree,, knots) { f(degree == 0){ B <- felse((x >= knots[]) & (x < knots[+1]), 1, 0) else { f((knots[degree+] - knots[]) == 0) { alpha1 <- 0 else { alpha1 <- (x - knots[])/(knots[degree+] - knots[]) f((knots[+degree+1] - knots[+1]) == 0) { alpha2 <- 0 else { alpha2 <- (knots[+degree+1] - x)/(knots[+degree+1] - knots[+1]) B <- alpha1*bass(x, (degree-1),, knots) + alpha2*bass(x, (degree-1), (+1), knots) return(b) bs <- functon(x, degree=3, nteror.knots=null, ntercept=false, Boundary.knots = c(0,1)) { f(mssng(x)) stop("you must provde x") f(degree < 1) stop("the splne degree must be at least 1") Boundary.knots <- sort(boundary.knots) nteror.knots.sorted <- NULL f(!s.null(nteror.knots)) nteror.knots.sorted <- sort(nteror.knots) knots <- c(rep(boundary.knots[1], (degree+1)), nteror.knots.sorted, rep(boundary.knots[2], (degree+1))) K <- length(nteror.knots) + degree + 1 B.mat <- matrx(0,length(x),k) for(j n 1:K) B.mat[,j] <- bass(x, degree, j, knots) f(any(x == Boundary.knots[2])) B.mat[x == Boundary.knots[2], K] <- 1 f(ntercept == FALSE) { return(b.mat[,-1]) else { return(b.mat) ## A smple llustraton that computes and plots the B-splne bases.
11 A PRIMER ON REGRESSION SPLINES 11 par(mfrow = c(2,1)) n < x <- seq(0, 1, length=n) B <- bs(x, degree=5, ntercept = TRUE, Boundary.knots=c(0, 1)) matplot(x, B, type="l", lwd=2) ## Next, smulate data then construct a regresson splne wth a ## prespecfed degree (n appled settngs we would want to choose ## the degree/knot vector usng a sound statstcal approach). dgp <- sn(2*p*x) y <- dgp + rnorm(n, sd=.1) model <- lm(y~b-1) plot(x, y, cex=.25, col="grey") lnes(x, ftted(model), lwd=2) lnes(x, dgp, col="red", lty=2)
BERNSTEIN POLYNOMIALS
On-Lne Geometrc Modelng Notes BERNSTEIN POLYNOMIALS Kenneth I. Joy Vsualzaton and Graphcs Research Group Department of Computer Scence Unversty of Calforna, Davs Overvew Polynomals are ncredbly useful
More informationRecurrence. 1 Definitions and main statements
Recurrence 1 Defntons and man statements Let X n, n = 0, 1, 2,... be a MC wth the state space S = (1, 2,...), transton probabltes p j = P {X n+1 = j X n = }, and the transton matrx P = (p j ),j S def.
More informationbenefit is 2, paid if the policyholder dies within the year, and probability of death within the year is ).
REVIEW OF RISK MANAGEMENT CONCEPTS LOSS DISTRIBUTIONS AND INSURANCE Loss and nsurance: When someone s subject to the rsk of ncurrng a fnancal loss, the loss s generally modeled usng a random varable or
More informationTHE METHOD OF LEAST SQUARES THE METHOD OF LEAST SQUARES
The goal: to measure (determne) an unknown quantty x (the value of a RV X) Realsaton: n results: y 1, y 2,..., y j,..., y n, (the measured values of Y 1, Y 2,..., Y j,..., Y n ) every result s encumbered
More information1 Example 1: Axis-aligned rectangles
COS 511: Theoretcal Machne Learnng Lecturer: Rob Schapre Lecture # 6 Scrbe: Aaron Schld February 21, 2013 Last class, we dscussed an analogue for Occam s Razor for nfnte hypothess spaces that, n conjuncton
More informationWhat is Candidate Sampling
What s Canddate Samplng Say we have a multclass or mult label problem where each tranng example ( x, T ) conssts of a context x a small (mult)set of target classes T out of a large unverse L of possble
More informationNPAR TESTS. One-Sample Chi-Square Test. Cell Specification. Observed Frequencies 1O i 6. Expected Frequencies 1EXP i 6
PAR TESTS If a WEIGHT varable s specfed, t s used to replcate a case as many tmes as ndcated by the weght value rounded to the nearest nteger. If the workspace requrements are exceeded and samplng has
More information8 Algorithm for Binary Searching in Trees
8 Algorthm for Bnary Searchng n Trees In ths secton we present our algorthm for bnary searchng n trees. A crucal observaton employed by the algorthm s that ths problem can be effcently solved when the
More informationWe are now ready to answer the question: What are the possible cardinalities for finite fields?
Chapter 3 Fnte felds We have seen, n the prevous chapters, some examples of fnte felds. For example, the resdue class rng Z/pZ (when p s a prme) forms a feld wth p elements whch may be dentfed wth the
More informationSupport Vector Machines
Support Vector Machnes Max Wellng Department of Computer Scence Unversty of Toronto 10 Kng s College Road Toronto, M5S 3G5 Canada wellng@cs.toronto.edu Abstract Ths s a note to explan support vector machnes.
More informationThe OC Curve of Attribute Acceptance Plans
The OC Curve of Attrbute Acceptance Plans The Operatng Characterstc (OC) curve descrbes the probablty of acceptng a lot as a functon of the lot s qualty. Fgure 1 shows a typcal OC Curve. 10 8 6 4 1 3 4
More informationRing structure of splines on triangulations
www.oeaw.ac.at Rng structure of splnes on trangulatons N. Vllamzar RICAM-Report 2014-48 www.rcam.oeaw.ac.at RING STRUCTURE OF SPLINES ON TRIANGULATIONS NELLY VILLAMIZAR Introducton For a trangulated regon
More informationCausal, Explanatory Forecasting. Analysis. Regression Analysis. Simple Linear Regression. Which is Independent? Forecasting
Causal, Explanatory Forecastng Assumes cause-and-effect relatonshp between system nputs and ts output Forecastng wth Regresson Analyss Rchard S. Barr Inputs System Cause + Effect Relatonshp The job of
More informationLuby s Alg. for Maximal Independent Sets using Pairwise Independence
Lecture Notes for Randomzed Algorthms Luby s Alg. for Maxmal Independent Sets usng Parwse Independence Last Updated by Erc Vgoda on February, 006 8. Maxmal Independent Sets For a graph G = (V, E), an ndependent
More information1. Measuring association using correlation and regression
How to measure assocaton I: Correlaton. 1. Measurng assocaton usng correlaton and regresson We often would lke to know how one varable, such as a mother's weght, s related to another varable, such as a
More informationPERRON FROBENIUS THEOREM
PERRON FROBENIUS THEOREM R. CLARK ROBINSON Defnton. A n n matrx M wth real entres m, s called a stochastc matrx provded () all the entres m satsfy 0 m, () each of the columns sum to one, m = for all, ()
More informationHow To Calculate The Accountng Perod Of Nequalty
Inequalty and The Accountng Perod Quentn Wodon and Shlomo Ytzha World Ban and Hebrew Unversty September Abstract Income nequalty typcally declnes wth the length of tme taen nto account for measurement.
More informationn + d + q = 24 and.05n +.1d +.25q = 2 { n + d + q = 24 (3) n + 2d + 5q = 40 (2)
MATH 16T Exam 1 : Part I (In-Class) Solutons 1. (0 pts) A pggy bank contans 4 cons, all of whch are nckels (5 ), dmes (10 ) or quarters (5 ). The pggy bank also contans a con of each denomnaton. The total
More informationCHAPTER 14 MORE ABOUT REGRESSION
CHAPTER 14 MORE ABOUT REGRESSION We learned n Chapter 5 that often a straght lne descrbes the pattern of a relatonshp between two quanttatve varables. For nstance, n Example 5.1 we explored the relatonshp
More informationLatent Class Regression. Statistics for Psychosocial Research II: Structural Models December 4 and 6, 2006
Latent Class Regresson Statstcs for Psychosocal Research II: Structural Models December 4 and 6, 2006 Latent Class Regresson (LCR) What s t and when do we use t? Recall the standard latent class model
More informationPSYCHOLOGICAL RESEARCH (PYC 304-C) Lecture 12
14 The Ch-squared dstrbuton PSYCHOLOGICAL RESEARCH (PYC 304-C) Lecture 1 If a normal varable X, havng mean µ and varance σ, s standardsed, the new varable Z has a mean 0 and varance 1. When ths standardsed
More informationQuantization Effects in Digital Filters
Quantzaton Effects n Dgtal Flters Dstrbuton of Truncaton Errors In two's complement representaton an exact number would have nfntely many bts (n general). When we lmt the number of bts to some fnte value
More informationSTATISTICAL DATA ANALYSIS IN EXCEL
Mcroarray Center STATISTICAL DATA ANALYSIS IN EXCEL Lecture 6 Some Advanced Topcs Dr. Petr Nazarov 14-01-013 petr.nazarov@crp-sante.lu Statstcal data analyss n Ecel. 6. Some advanced topcs Correcton for
More informationv a 1 b 1 i, a 2 b 2 i,..., a n b n i.
SECTION 8.4 COMPLEX VECTOR SPACES AND INNER PRODUCTS 455 8.4 COMPLEX VECTOR SPACES AND INNER PRODUCTS All the vector spaces we have studed thus far n the text are real vector spaces snce the scalars are
More informationTHE DISTRIBUTION OF LOAN PORTFOLIO VALUE * Oldrich Alfons Vasicek
HE DISRIBUION OF LOAN PORFOLIO VALUE * Oldrch Alfons Vascek he amount of captal necessary to support a portfolo of debt securtes depends on the probablty dstrbuton of the portfolo loss. Consder a portfolo
More informationForecasting the Direction and Strength of Stock Market Movement
Forecastng the Drecton and Strength of Stock Market Movement Jngwe Chen Mng Chen Nan Ye cjngwe@stanford.edu mchen5@stanford.edu nanye@stanford.edu Abstract - Stock market s one of the most complcated systems
More informationPRACTICE 1: MUTUAL FUNDS EVALUATION USING MATLAB.
PRACTICE 1: MUTUAL FUNDS EVALUATION USING MATLAB. INDEX 1. Load data usng the Edtor wndow and m-fle 2. Learnng to save results from the Edtor wndow. 3. Computng the Sharpe Rato 4. Obtanng the Treynor Rato
More informationCHOLESTEROL REFERENCE METHOD LABORATORY NETWORK. Sample Stability Protocol
CHOLESTEROL REFERENCE METHOD LABORATORY NETWORK Sample Stablty Protocol Background The Cholesterol Reference Method Laboratory Network (CRMLN) developed certfcaton protocols for total cholesterol, HDL
More informationwhere the coordinates are related to those in the old frame as follows.
Chapter 2 - Cartesan Vectors and Tensors: Ther Algebra Defnton of a vector Examples of vectors Scalar multplcaton Addton of vectors coplanar vectors Unt vectors A bass of non-coplanar vectors Scalar product
More informationAn Alternative Way to Measure Private Equity Performance
An Alternatve Way to Measure Prvate Equty Performance Peter Todd Parlux Investment Technology LLC Summary Internal Rate of Return (IRR) s probably the most common way to measure the performance of prvate
More informationModule 2 LOSSLESS IMAGE COMPRESSION SYSTEMS. Version 2 ECE IIT, Kharagpur
Module LOSSLESS IMAGE COMPRESSION SYSTEMS Lesson 3 Lossless Compresson: Huffman Codng Instructonal Objectves At the end of ths lesson, the students should be able to:. Defne and measure source entropy..
More informationLogistic Regression. Lecture 4: More classifiers and classes. Logistic regression. Adaboost. Optimization. Multiple class classification
Lecture 4: More classfers and classes C4B Machne Learnng Hlary 20 A. Zsserman Logstc regresson Loss functons revsted Adaboost Loss functons revsted Optmzaton Multple class classfcaton Logstc Regresson
More informationExtending Probabilistic Dynamic Epistemic Logic
Extendng Probablstc Dynamc Epstemc Logc Joshua Sack May 29, 2008 Probablty Space Defnton A probablty space s a tuple (S, A, µ), where 1 S s a set called the sample space. 2 A P(S) s a σ-algebra: a set
More informationImplementation of Deutsch's Algorithm Using Mathcad
Implementaton of Deutsch's Algorthm Usng Mathcad Frank Roux The followng s a Mathcad mplementaton of Davd Deutsch's quantum computer prototype as presented on pages - n "Machnes, Logc and Quantum Physcs"
More information8.5 UNITARY AND HERMITIAN MATRICES. The conjugate transpose of a complex matrix A, denoted by A*, is given by
6 CHAPTER 8 COMPLEX VECTOR SPACES 5. Fnd the kernel of the lnear transformaton gven n Exercse 5. In Exercses 55 and 56, fnd the mage of v, for the ndcated composton, where and are gven by the followng
More informationRegression Models for a Binary Response Using EXCEL and JMP
SEMATECH 997 Statstcal Methods Symposum Austn Regresson Models for a Bnary Response Usng EXCEL and JMP Davd C. Trndade, Ph.D. STAT-TECH Consultng and Tranng n Appled Statstcs San Jose, CA Topcs Practcal
More informationCalculation of Sampling Weights
Perre Foy Statstcs Canada 4 Calculaton of Samplng Weghts 4.1 OVERVIEW The basc sample desgn used n TIMSS Populatons 1 and 2 was a two-stage stratfed cluster desgn. 1 The frst stage conssted of a sample
More informationLecture 3: Force of Interest, Real Interest Rate, Annuity
Lecture 3: Force of Interest, Real Interest Rate, Annuty Goals: Study contnuous compoundng and force of nterest Dscuss real nterest rate Learn annuty-mmedate, and ts present value Study annuty-due, and
More informationHedging Interest-Rate Risk with Duration
FIXED-INCOME SECURITIES Chapter 5 Hedgng Interest-Rate Rsk wth Duraton Outlne Prcng and Hedgng Prcng certan cash-flows Interest rate rsk Hedgng prncples Duraton-Based Hedgng Technques Defnton of duraton
More informationGeneralizing the degree sequence problem
Mddlebury College March 2009 Arzona State Unversty Dscrete Mathematcs Semnar The degree sequence problem Problem: Gven an nteger sequence d = (d 1,...,d n ) determne f there exsts a graph G wth d as ts
More informationLeast Squares Fitting of Data
Least Squares Fttng of Data Davd Eberly Geoetrc Tools, LLC http://www.geoetrctools.co/ Copyrght c 1998-2016. All Rghts Reserved. Created: July 15, 1999 Last Modfed: January 5, 2015 Contents 1 Lnear Fttng
More informationNumerical Methods 數 值 方 法 概 說. Daniel Lee. Nov. 1, 2006
Numercal Methods 數 值 方 法 概 說 Danel Lee Nov. 1, 2006 Outlnes Lnear system : drect, teratve Nonlnear system : Newton-lke Interpolatons : polys, splnes, trg polys Approxmatons (I) : orthogonal polys Approxmatons
More informationStatistical Methods to Develop Rating Models
Statstcal Methods to Develop Ratng Models [Evelyn Hayden and Danel Porath, Österrechsche Natonalbank and Unversty of Appled Scences at Manz] Source: The Basel II Rsk Parameters Estmaton, Valdaton, and
More informationRate-Based Daily Arrival Process Models with Application to Call Centers
Submtted to Operatons Research manuscrpt (Please, provde the manuscrpt number!) Authors are encouraged to submt new papers to INFORMS journals by means of a style fle template, whch ncludes the journal
More informationBrigid Mullany, Ph.D University of North Carolina, Charlotte
Evaluaton And Comparson Of The Dfferent Standards Used To Defne The Postonal Accuracy And Repeatablty Of Numercally Controlled Machnng Center Axes Brgd Mullany, Ph.D Unversty of North Carolna, Charlotte
More informationHow To Understand The Results Of The German Meris Cloud And Water Vapour Product
Ttel: Project: Doc. No.: MERIS level 3 cloud and water vapour products MAPP MAPP-ATBD-ClWVL3 Issue: 1 Revson: 0 Date: 9.12.1998 Functon Name Organsaton Sgnature Date Author: Bennartz FUB Preusker FUB Schüller
More informationProduction. 2. Y is closed A set is closed if it contains its boundary. We need this for the solution existence in the profit maximization problem.
Producer Theory Producton ASSUMPTION 2.1 Propertes of the Producton Set The producton set Y satsfes the followng propertes 1. Y s non-empty If Y s empty, we have nothng to talk about 2. Y s closed A set
More informationUsage of LCG/CLCG numbers for electronic gambling applications
Usage of LCG/CLCG numbers for electronc gamblng applcatons Anders Knutsson Smovts Consultng, Wenner-Gren Center, Sveavägen 166, 113 46 Stockholm, Sweden anders.knutsson@smovts.com Abstract. Several attacks
More informationCS 2750 Machine Learning. Lecture 3. Density estimation. CS 2750 Machine Learning. Announcements
Lecture 3 Densty estmaton Mlos Hauskrecht mlos@cs.ptt.edu 5329 Sennott Square Next lecture: Matlab tutoral Announcements Rules for attendng the class: Regstered for credt Regstered for audt (only f there
More informationMARKET SHARE CONSTRAINTS AND THE LOSS FUNCTION IN CHOICE BASED CONJOINT ANALYSIS
MARKET SHARE CONSTRAINTS AND THE LOSS FUNCTION IN CHOICE BASED CONJOINT ANALYSIS Tmothy J. Glbrde Assstant Professor of Marketng 315 Mendoza College of Busness Unversty of Notre Dame Notre Dame, IN 46556
More informationFormulating & Solving Integer Problems Chapter 11 289
Formulatng & Solvng Integer Problems Chapter 11 289 The Optonal Stop TSP If we drop the requrement that every stop must be vsted, we then get the optonal stop TSP. Ths mght correspond to a ob sequencng
More informationFast degree elevation and knot insertion for B-spline curves
Computer Aded Geometrc Desgn 22 (2005) 183 197 www.elsever.com/locate/cagd Fast degree elevaton and knot nserton for B-splne curves Q-Xng Huang a,sh-mnhu a,, Ralph R. Martn b a Department of Computer Scence
More informationConstruction Rules for Morningstar Canada Target Dividend Index SM
Constructon Rules for Mornngstar Canada Target Dvdend Index SM Mornngstar Methodology Paper October 2014 Verson 1.2 2014 Mornngstar, Inc. All rghts reserved. The nformaton n ths document s the property
More informationYIELD CURVE FITTING 2.0 Constructing Bond and Money Market Yield Curves using Cubic B-Spline and Natural Cubic Spline Methodology.
YIELD CURVE FITTING 2.0 Constructng Bond and Money Market Yeld Curves usng Cubc B-Splne and Natural Cubc Splne Methodology Users Manual YIELD CURVE FITTING 2.0 Users Manual Authors: Zhuosh Lu, Moorad Choudhry
More informationOnline Appendix for Forecasting the Equity Risk Premium: The Role of Technical Indicators
Onlne Appendx for Forecastng the Equty Rsk Premum: The Role of Techncal Indcators Chrstopher J. Neely Federal Reserve Bank of St. Lous neely@stls.frb.org Davd E. Rapach Sant Lous Unversty rapachde@slu.edu
More informationThe Greedy Method. Introduction. 0/1 Knapsack Problem
The Greedy Method Introducton We have completed data structures. We now are gong to look at algorthm desgn methods. Often we are lookng at optmzaton problems whose performance s exponental. For an optmzaton
More informationThe Development of Web Log Mining Based on Improve-K-Means Clustering Analysis
The Development of Web Log Mnng Based on Improve-K-Means Clusterng Analyss TngZhong Wang * College of Informaton Technology, Luoyang Normal Unversty, Luoyang, 471022, Chna wangtngzhong2@sna.cn Abstract.
More informationHow Sets of Coherent Probabilities May Serve as Models for Degrees of Incoherence
1 st Internatonal Symposum on Imprecse Probabltes and Ther Applcatons, Ghent, Belgum, 29 June 2 July 1999 How Sets of Coherent Probabltes May Serve as Models for Degrees of Incoherence Mar J. Schervsh
More informationL10: Linear discriminants analysis
L0: Lnear dscrmnants analyss Lnear dscrmnant analyss, two classes Lnear dscrmnant analyss, C classes LDA vs. PCA Lmtatons of LDA Varants of LDA Other dmensonalty reducton methods CSCE 666 Pattern Analyss
More informationANALYZING THE RELATIONSHIPS BETWEEN QUALITY, TIME, AND COST IN PROJECT MANAGEMENT DECISION MAKING
ANALYZING THE RELATIONSHIPS BETWEEN QUALITY, TIME, AND COST IN PROJECT MANAGEMENT DECISION MAKING Matthew J. Lberatore, Department of Management and Operatons, Vllanova Unversty, Vllanova, PA 19085, 610-519-4390,
More information+ + + - - This circuit than can be reduced to a planar circuit
MeshCurrent Method The meshcurrent s analog of the nodeoltage method. We sole for a new set of arables, mesh currents, that automatcally satsfy KCLs. As such, meshcurrent method reduces crcut soluton to
More informationSection 5.3 Annuities, Future Value, and Sinking Funds
Secton 5.3 Annutes, Future Value, and Snkng Funds Ordnary Annutes A sequence of equal payments made at equal perods of tme s called an annuty. The tme between payments s the payment perod, and the tme
More informationSingle and multiple stage classifiers implementing logistic discrimination
Sngle and multple stage classfers mplementng logstc dscrmnaton Hélo Radke Bttencourt 1 Dens Alter de Olvera Moraes 2 Vctor Haertel 2 1 Pontfíca Unversdade Católca do Ro Grande do Sul - PUCRS Av. Ipranga,
More informationMinimal Coding Network With Combinatorial Structure For Instantaneous Recovery From Edge Failures
Mnmal Codng Network Wth Combnatoral Structure For Instantaneous Recovery From Edge Falures Ashly Joseph 1, Mr.M.Sadsh Sendl 2, Dr.S.Karthk 3 1 Fnal Year ME CSE Student Department of Computer Scence Engneerng
More informationREGULAR MULTILINEAR OPERATORS ON C(K) SPACES
REGULAR MULTILINEAR OPERATORS ON C(K) SPACES FERNANDO BOMBAL AND IGNACIO VILLANUEVA Abstract. The purpose of ths paper s to characterze the class of regular contnuous multlnear operators on a product of
More informationNonbinary Quantum Error-Correcting Codes from Algebraic Curves
Nonbnary Quantum Error-Correctng Codes from Algebrac Curves Jon-Lark Km and Judy Walker Department of Mathematcs Unversty of Nebraska-Lncoln, Lncoln, NE 68588-0130 USA e-mal: {jlkm, jwalker}@math.unl.edu
More informationAn Interest-Oriented Network Evolution Mechanism for Online Communities
An Interest-Orented Network Evoluton Mechansm for Onlne Communtes Cahong Sun and Xaopng Yang School of Informaton, Renmn Unversty of Chna, Bejng 100872, P.R. Chna {chsun,yang}@ruc.edu.cn Abstract. Onlne
More information1 De nitions and Censoring
De ntons and Censorng. Survval Analyss We begn by consderng smple analyses but we wll lead up to and take a look at regresson on explanatory factors., as n lnear regresson part A. The mportant d erence
More informationJ. Parallel Distrib. Comput.
J. Parallel Dstrb. Comput. 71 (2011) 62 76 Contents lsts avalable at ScenceDrect J. Parallel Dstrb. Comput. journal homepage: www.elsever.com/locate/jpdc Optmzng server placement n dstrbuted systems n
More informationSection 5.4 Annuities, Present Value, and Amortization
Secton 5.4 Annutes, Present Value, and Amortzaton Present Value In Secton 5.2, we saw that the present value of A dollars at nterest rate per perod for n perods s the amount that must be deposted today
More informationLoop Parallelization
- - Loop Parallelzaton C-52 Complaton steps: nested loops operatng on arrays, sequentell executon of teraton space DECLARE B[..,..+] FOR I :=.. FOR J :=.. I B[I,J] := B[I-,J]+B[I-,J-] ED FOR ED FOR analyze
More informationJoe Pimbley, unpublished, 2005. Yield Curve Calculations
Joe Pmbley, unpublshed, 005. Yeld Curve Calculatons Background: Everythng s dscount factors Yeld curve calculatons nclude valuaton of forward rate agreements (FRAs), swaps, nterest rate optons, and forward
More informationExhaustive Regression. An Exploration of Regression-Based Data Mining Techniques Using Super Computation
Exhaustve Regresson An Exploraton of Regresson-Based Data Mnng Technques Usng Super Computaton Antony Daves, Ph.D. Assocate Professor of Economcs Duquesne Unversty Pttsburgh, PA 58 Research Fellow The
More informationRisk-based Fatigue Estimate of Deep Water Risers -- Course Project for EM388F: Fracture Mechanics, Spring 2008
Rsk-based Fatgue Estmate of Deep Water Rsers -- Course Project for EM388F: Fracture Mechancs, Sprng 2008 Chen Sh Department of Cvl, Archtectural, and Envronmental Engneerng The Unversty of Texas at Austn
More informationAn Inductive Fuzzy Classification Approach applied to Individual Marketing
An Inductve Fuzzy Classfcaton Approach appled to Indvdual Marketng Mchael Kaufmann, Andreas Meer Abstract A data mnng methodology for an nductve fuzzy classfcaton s ntroduced. The nducton step s based
More informationGeneral Iteration Algorithm for Classification Ratemaking
General Iteraton Algorthm for Classfcaton Ratemakng by Luyang Fu and Cheng-sheng eter Wu ABSTRACT In ths study, we propose a flexble and comprehensve teraton algorthm called general teraton algorthm (GIA)
More informationPricing Multi-Asset Cross Currency Options
CIRJE-F-844 Prcng Mult-Asset Cross Currency Optons Kenchro Shraya Graduate School of Economcs, Unversty of Tokyo Akhko Takahash Unversty of Tokyo March 212; Revsed n September, October and November 212
More informationProject Networks With Mixed-Time Constraints
Project Networs Wth Mxed-Tme Constrants L Caccetta and B Wattananon Western Australan Centre of Excellence n Industral Optmsaton (WACEIO) Curtn Unversty of Technology GPO Box U1987 Perth Western Australa
More informationProduct-Form Stationary Distributions for Deficiency Zero Chemical Reaction Networks
Bulletn of Mathematcal Bology (21 DOI 1.17/s11538-1-9517-4 ORIGINAL ARTICLE Product-Form Statonary Dstrbutons for Defcency Zero Chemcal Reacton Networks Davd F. Anderson, Gheorghe Cracun, Thomas G. Kurtz
More informationDiagnostic Tests of Cross Section Independence for Nonlinear Panel Data Models
DISCUSSION PAPER SERIES IZA DP No. 2756 Dagnostc ests of Cross Secton Independence for Nonlnear Panel Data Models Cheng Hsao M. Hashem Pesaran Andreas Pck Aprl 2007 Forschungsnsttut zur Zukunft der Arbet
More informationNMT EE 589 & UNM ME 482/582 ROBOT ENGINEERING. Dr. Stephen Bruder NMT EE 589 & UNM ME 482/582
NMT EE 589 & UNM ME 482/582 ROBOT ENGINEERING Dr. Stephen Bruder NMT EE 589 & UNM ME 482/582 7. Root Dynamcs 7.2 Intro to Root Dynamcs We now look at the forces requred to cause moton of the root.e. dynamcs!!
More informationRealistic Image Synthesis
Realstc Image Synthess - Combned Samplng and Path Tracng - Phlpp Slusallek Karol Myszkowsk Vncent Pegoraro Overvew: Today Combned Samplng (Multple Importance Samplng) Renderng and Measurng Equaton Random
More informationSurvival analysis methods in Insurance Applications in car insurance contracts
Survval analyss methods n Insurance Applcatons n car nsurance contracts Abder OULIDI 1 Jean-Mare MARION 2 Hervé GANACHAUD 3 Abstract In ths wor, we are nterested n survval models and ther applcatons on
More information1. Fundamentals of probability theory 2. Emergence of communication traffic 3. Stochastic & Markovian Processes (SP & MP)
6.3 / -- Communcaton Networks II (Görg) SS20 -- www.comnets.un-bremen.de Communcaton Networks II Contents. Fundamentals of probablty theory 2. Emergence of communcaton traffc 3. Stochastc & Markovan Processes
More informationIDENTIFICATION AND CORRECTION OF A COMMON ERROR IN GENERAL ANNUITY CALCULATIONS
IDENTIFICATION AND CORRECTION OF A COMMON ERROR IN GENERAL ANNUITY CALCULATIONS Chrs Deeley* Last revsed: September 22, 200 * Chrs Deeley s a Senor Lecturer n the School of Accountng, Charles Sturt Unversty,
More informationAbstract. Clustering ensembles have emerged as a powerful method for improving both the
Clusterng Ensembles: {topchyal, Models jan, of punch}@cse.msu.edu Consensus and Weak Parttons * Alexander Topchy, Anl K. Jan, and Wllam Punch Department of Computer Scence and Engneerng, Mchgan State Unversty
More informationInter-Ing 2007. INTERDISCIPLINARITY IN ENGINEERING SCIENTIFIC INTERNATIONAL CONFERENCE, TG. MUREŞ ROMÂNIA, 15-16 November 2007.
Inter-Ing 2007 INTERDISCIPLINARITY IN ENGINEERING SCIENTIFIC INTERNATIONAL CONFERENCE, TG. MUREŞ ROMÂNIA, 15-16 November 2007. UNCERTAINTY REGION SIMULATION FOR A SERIAL ROBOT STRUCTURE MARIUS SEBASTIAN
More informationChapter XX More advanced approaches to the analysis of survey data. Gad Nathan Hebrew University Jerusalem, Israel. Abstract
Household Sample Surveys n Developng and Transton Countres Chapter More advanced approaches to the analyss of survey data Gad Nathan Hebrew Unversty Jerusalem, Israel Abstract In the present chapter, we
More information) of the Cell class is created containing information about events associated with the cell. Events are added to the Cell instance
Calbraton Method Instances of the Cell class (one nstance for each FMS cell) contan ADC raw data and methods assocated wth each partcular FMS cell. The calbraton method ncludes event selecton (Class Cell
More informationA Probabilistic Theory of Coherence
A Probablstc Theory of Coherence BRANDEN FITELSON. The Coherence Measure C Let E be a set of n propostons E,..., E n. We seek a probablstc measure C(E) of the degree of coherence of E. Intutvely, we want
More informationSection C2: BJT Structure and Operational Modes
Secton 2: JT Structure and Operatonal Modes Recall that the semconductor dode s smply a pn juncton. Dependng on how the juncton s based, current may easly flow between the dode termnals (forward bas, v
More informationA machine vision approach for detecting and inspecting circular parts
A machne vson approach for detectng and nspectng crcular parts Du-Mng Tsa Machne Vson Lab. Department of Industral Engneerng and Management Yuan-Ze Unversty, Chung-L, Tawan, R.O.C. E-mal: edmtsa@saturn.yzu.edu.tw
More informationEfficient Project Portfolio as a tool for Enterprise Risk Management
Effcent Proect Portfolo as a tool for Enterprse Rsk Management Valentn O. Nkonov Ural State Techncal Unversty Growth Traectory Consultng Company January 5, 27 Effcent Proect Portfolo as a tool for Enterprse
More informationTexas Instruments 30X IIS Calculator
Texas Instruments 30X IIS Calculator Keystrokes for the TI-30X IIS are shown for a few topcs n whch keystrokes are unque. Start by readng the Quk Start secton. Then, before begnnng a specfc unt of the
More informationCan Auto Liability Insurance Purchases Signal Risk Attitude?
Internatonal Journal of Busness and Economcs, 2011, Vol. 10, No. 2, 159-164 Can Auto Lablty Insurance Purchases Sgnal Rsk Atttude? Chu-Shu L Department of Internatonal Busness, Asa Unversty, Tawan Sheng-Chang
More information7.5. Present Value of an Annuity. Investigate
7.5 Present Value of an Annuty Owen and Anna are approachng retrement and are puttng ther fnances n order. They have worked hard and nvested ther earnngs so that they now have a large amount of money on
More informationSecure Network Coding Over the Integers
Secure Network Codng Over the Integers Rosaro Gennaro Jonathan Katz Hugo Krawczyk Tal Rabn Abstract Network codng has receved sgnfcant attenton n the networkng communty for ts potental to ncrease throughput
More informationThe Racial and Gender Interest Rate Gap. in Small Business Lending: Improved Estimates Using Matching Methods*
The Racal and Gender Interest Rate Gap n Small Busness Lendng: Improved Estmates Usng Matchng Methods* Yue Hu and Long Lu Department of Economcs Unversty of Texas at San Antono Jan Ondrch and John Ynger
More informationNON-CONSTANT SUM RED-AND-BLACK GAMES WITH BET-DEPENDENT WIN PROBABILITY FUNCTION LAURA PONTIGGIA, University of the Sciences in Philadelphia
To appear n Journal o Appled Probablty June 2007 O-COSTAT SUM RED-AD-BLACK GAMES WITH BET-DEPEDET WI PROBABILITY FUCTIO LAURA POTIGGIA, Unversty o the Scences n Phladelpha Abstract In ths paper we nvestgate
More informationHow To Know The Components Of Mean Squared Error Of Herarchcal Estmator S
S C H E D A E I N F O R M A T I C A E VOLUME 0 0 On Mean Squared Error of Herarchcal Estmator Stans law Brodowsk Faculty of Physcs, Astronomy, and Appled Computer Scence, Jagellonan Unversty, Reymonta
More information