CS 2750 Machine Learning. Lecture 7. Linear regression. CS 2750 Machine Learning. Linear regression. is a linear combination of input components x
|
|
- Aubrie Craig
- 5 years ago
- Views:
Transcription
1 CS 75 Mache Learg Lecture 7 Lear regresso Mlos Hauskrecht mlos@cs.ptt.eu 539 Seott Square CS 75 Mache Learg Lear regresso Fucto f : X Y s a lear combato of put compoets f k - parameters eghts Bas term f Iput vector CS 75 Mache Learg
2 Lear regresso. Error. Data: D Fucto: f We oul lke to have f for all.. Error fucto measures ho much our prectos evate from the esre asers Mea-square error.. Learg: We at to f the eghts mmzg the error! f CS 75 Mache Learg Lear regresso. Eample mesoal put CS 75 Mache Learg
3 Lear regresso. Eample. mesoal put CS 75 Mache Learg Lear regresso. Optmzato. We at the eghts mmzg the error f.. For the optmal set of parameters ervatves of the error th respect to each parameter must be Vector of ervatves: gra.. CS 75 Mache Learg 3
4 4 CS 75 Mache Learg Lear regresso. Optmzato. efes a set of equatos gra CS 75 Mache Learg Solvg lear regresso B rearragg the terms e get a sstem of lear equatos th + ukos A b
5 5 CS 75 Mache Learg Solvg lear regresso he optmal set of eghts satsfes: Leas to a sstem of lear equatos SLE th + ukos of the form Soluto to SLE: matr verso A b b A CS 75 Mache Learg Graet escet soluto Goal: the eght optmzato the lear regresso moel A alteratve to SLE soluto: Graet escet Iea: Aust eghts the recto that mproves the Error he graet tells us hat s the rght recto - a learg rate scales the graet chages Error.. f Error
6 Graet escet metho Desce usg the graet formato Error Error * * Drecto of the escet Chage the value of accorg to the graet Error CS 75 Mache Learg Graet escet metho Error Error * * Ne value of the parameter Error * * - a learg rate scales the graet chages CS 75 Mache Learg For all 6
7 Graet escet metho Iteratvel approaches the optmum of the Error fucto Error 3 CS 75 Mache Learg Ole graet algorthm he error fucto s efe for the hole ataset D Error f error for a sample.. ole Error f Ole graet metho: chages eghts after ever sample Error vector form: D Error - Learg rate that epes o the umber of upates CS 75 Mache Learg 7
8 Ole graet metho Lear moel f O-le error ole Error f O-le algorthm: geerates a sequece of ole upates -th upate step th : -th eght: D Error f Fe learg rate: - Use a small costat C Aeale learg rate: - Grauall rescales chages CS 75 Mache Learg Ole regresso algorthm Ole-lear-regresso D umber of teratos Italze eghts for =:: umber of teratos o select a ata pot D from D set learg rate upate eght vector f e for retur eghts Avatages: ver eas to mplemet cotuous ata streams CS 75 Mache Learg 8
9 O-le learg. Eample CS 75 Mache Learg Practcal cocers: Iput ormalzato Iput ormalzato makes the ata var roughl o the same scale. Ca make a huge fferece o-le learg Assume o-le upate elta rule for to eghts k: f = k k f k Chage epes o the magtue of the put For puts th a large magtue the chage the eght s huge: chages to the puts th hgh magtue sproportoal as f the put as more mportat CS 75 Mache Learg 9
10 Iput ormalzato Iput ormalzato: Soluto to the problem of fferet scales Makes all puts var the same rage arou Ne put: ~ More comple ormalzato approach ca be apple he e at to process ata th correlatos Smlarl e ca reormalze outputs CS 75 Mache Learg Etesos of smple lear moel Replace puts to lear uts th feature bass fuctos to moel oleartes f m f - a arbtrar fucto of m m he same techques as before to lear the eghts CS 75 Mache Learg
11 Atve lear moels Moels lear the parameters e at to ft m f k k k... m - parameters... m - feature or bass fuctos Bass fuctos eamples: a hgher orer polomal oe-mesoal put 3 3 Multmesoal quaratc Other tpes of bass fuctos s cos CS 75 Mache Learg Fttg atve lear moels Error fucto /.. f Assume: φ.. Leas to a sstem of m lear equatos f φ m m m Ca be solve eactl lke the lear case CS 75 Mache Learg
12 Eample. Regresso th polomals. Regresso th polomals of egree m Data pots: pars of Feature fuctos: m feature fuctos m Fucto to lear: m f m m m m CS 75 Mache Learg Learg th feature fuctos Fucto to lear: f O le graet upate for the <> par f k f Graet upates are of the same form as the lear a logstc regresso moels CS 75 Mache Learg
13 Eample. Regresso th polomals. Eample: Regresso th polomals of egree m m f O le upate for <> par f m f CS 75 Mache Learg Multmesoal atve moel eample CS 75 Mache Learg 3
14 Multmesoal atve moel eample CS 75 Mache Learg Statstcal moel of regresso A geeratve moel: f f s a etermstc fucto s a raom ose represets thgs e caot capture th f e.g. ~ N Assume f s a lear moel a ~ N he: f E moels the mea of outputs for a the ose moels evatos from the mea he moel efes the cotoal est of gve p ep f CS 75 Mache Learg 4
15 5 CS 75 Mache Learg ML estmato of the parameters lkelhoo of prectos = the probablt of observg outputs D gve Mamum lkelhoo estmato of parameters parameters mamzg the lkelhoo of prectos Log-lkelhoo trck for the ML optmzato Mamzg the log-lkelhoo s equvalet to mamzg the lkelhoo p D L p * ma arg p D L D l log log CS 75 Mache Learg ML estmato of the parameters Usg cotoal est We ca rerte the log-lkelhoo as Mamzg th regar to s equvalet to mmzg square error fuctos p D L D l log log c f p log C f ] ep[ f p
16 ML estmato of parameters Crtera base o mea squares error fucto a the log lkelhoo of the output are relate ole log p c We ko ho to optmze parameters the same approach as use for the least squares ft But hat s the ML estmate of the varace of the ose? Mamze l D th respect to varace ˆ f = mea square precto error for the best prector * CS 75 Mache Learg Regularze lear regresso If the umber of parameters s large relatve to the umber of ata pots use to tra the moel e face the threat of overft geeralzato error of the moel goes up he precto accurac ca be ofte mprove b settg some coeffcets to zero Icreases the bas reuces the varace of estmates Solutos: Subset selecto Rge regresso Lasso regresso Prcpal compoet regresso Net: rge regresso 6
17 Rge regresso Error fucto for the staar least squares estmates:.. * We seek: arg m Rge regresso: Where.... a What oes the e error fucto o? Rge regresso Staar regresso: Rge regresso:.. L.. pealzes o-zero eghts th the cost L proportoal to a shrkage coeffcet If a put attrbute has a small effect o mprovg the error fucto t s shut o b the pealt term Icluso of a shrkage pealt s ofte referre to as regularzato 7
18 Regularze lear regresso Ho to solve the least squares problem f the error fucto s erche b the regularzato term? Aser: he soluto to the optmal set of eghts s obtae aga b solvg a set of lear equato. Staar lear regresso: Soluto: * X X X Regularze lear regresso: here X s a matr th ros correspog to eamples a colums to puts * I X X X Lasso regresso Staar regresso: Lasso regresso:.... pealzes o-zero eghts th the cost proportoal to Lasso regresso s more aggressve tha the rge regresso zerog the eghts Lasso + rge regularzato combe: Elastc et regularzato 8
CS 1675 Intro to Machine Learning Lecture 9. Linear regression. Supervised learning. a set of n examples
CS 675 Itro to Mache Learg Lecture 9 Lear regresso Mlos Hauskrecht mlos@cs.ptt.eu 59 Seott Square Supervse learg Data: D { D D.. D} a set of eamples D s a put vector of sze s the esre output gve b a teacher
More informationLinear regression II
CS 75 Mache Learg Lecture 9 Lear regresso II Mlos Hauskrecht mlos@cs.ptt.eu 539 Seott Square Lear regresso Fucto f : X Y Y s a lear combato of put compoets f ( w w w w w w, w, w k - parameters (weghts
More informationDensity estimation II.
Lecture 5 esty estmato II. Mlos Hausrecht mlos@cs.tt.eu 539 Seott Square Outle Outle: esty estmato: Bomal strbuto Multomal strbuto ormal strbuto oetal famly ata: esty estmato {.. } a vector of attrbute
More informationCHAPTER 8. r E( r ) m e. Reduces the number of inputs for diversification. Easier for security analysts to specialize
CHATE 8 Idex odels cgra-hll/ir Copyrght 0 by The cgra-hll Compaes, Ic. All rghts reserved. 8- Advatages of the Sgle Idex odel educes the umber of puts for dversfcato Easer for securty aalysts to specalze
More informationOverview. Linear Models Connectionist and Statistical Language Processing. Numeric Prediction. Example
Overvew Lear Models Coectost ad Statstcal Laguage Processg Frak Keller keller@col.u-sb.de Computerlgustk Uverstät des Saarlades classfcato vs. umerc predcto lear regresso least square estmato evaluatg
More informationGross Product Simulation with pooling of Linear and Nonlinear Regression Models
Gross Prouct Smulato wth poolg of Lear a Nolear Regresso Moels Ahma Flah, Abbas Abalmuhse, Ebtsam Abulah, Sr Ramaswam Uverst of Arkasas at Lttle Rock Lttle Rock, AR 704, USA {aflah, akabalmuh, ekabulah,
More informationMeasures of Dispersion
Chapter IV Meaure of Dpero R. 4.. The meaure of locato cate the geeral magtue of the ata a locate oly the cetre of a trbuto. They o ot etablh the egree of varablty or the prea out or catter of the vual
More informationSCEA CERTIFICATION EXAM: PRACTICE QUESTIONS AND STUDY AID
SCEA CERTIFICATION EAM: PRACTICE QUESTIONS AND STUDY AID Lear Regresso Formulas Cheat Sheet You ma use the followg otes o lear regresso to work eam questos. Let be a depedet varable ad be a depedet varable
More informationDeriving & Understanding the Variance Formulas
Dervg & Uderstadg the Varace Formulas Ma H. Farrell BUS 400 August 28, 205 The purpose of ths hadout s to derve the varace formulas that we dscussed class ad show why take the form they do. I class we
More informationThe Complexity of General Equilibrium
Prof. Ja Bhattachara Eco --Sprg 200 Welfare Propertes of Market Outcomes Last tme, we covered equlbrum oe market partal equlbrum. We foud that uder perfect competto, the equlbrum prce ad quatt mamzed the
More informationPortfolio Optimization: MAD vs. Markowitz
Rose-Hulma Udergraduate Mathematcs Joural Volume 6 Issue 2 Artcle 3 Portfolo Optmzato: MAD vs. Markowtz Beth Bower College of Wllam ad Mary, bebowe@wm.edu Pamela Wetz Mllersvlle Uversty, pamela037@hotmal.com
More informationGene Expression Data Analysis (II) statistical issues in spotted arrays
STATC4 Sprg 005 Lecture Data ad fgures are from Wg Wog s computatoal bology course at Harvard Gee Expresso Data Aalyss (II) statstcal ssues spotted arrays Below shows part of a result fle from mage aalyss
More informationThe Firm. The Firm. Maximizing Profits. Decisions. ECON 370: Microeconomic Theory Summer 2004 Rice University Stanley Gilbert
The Frm The Frm ECON 370: Mcroecoomc Theory Summer 004 Rce Uversty Staley Glbert A Frm s a mechasm for covertg labor, captal ad raw materals to desrable goods A frm s owed by cosumers ad operated for the
More informationOptimal Reliability Allocation
Optmal Relablty Allocato Yashwat K. Malaya malaya@cs.colostate.edu Departmet of Computer Scece Colorado State Uversty Relablty Allocato Problem Allocato the relablty values to subsystems to mmze the total
More informationProbability and Statistical Methods. Chapter 8 Fundamental Sampling Distributions
Math 3 Probablty ad Statstcal Methods Chapter 8 Fudametal Samplg Dstrbutos Samplg Dstrbutos I the process of makg a ferece from a sample to a populato we usually calculate oe or more statstcs, such as
More informationProbability and Statistical Methods. Chapter 8 Fundamental Sampling Distributions
Math 3 Probablty ad Statstcal Methods Chapter 8 Fudametal Samplg Dstrbutos Samplg Dstrbutos I the process of makg a ferece from a sample to a populato we usually calculate oe or more statstcs, such as
More informationVariable weight combined forecast of China s energy demand based on grey model and BP neural network
Avalable ole www.jocpr.com Joural of Chemcal ad Pharmaceutcal Research, 2014, 6(4):303-308 Research Artcle ISSN : 0975-7384 CODEN(USA) : JCPRC5 Varable weght combed forecast of Cha s eergy demad based
More information8.0% E(R) 6.0% Lend. Borrow 4.0% 2.0% rf rf 0.0% 0.0% 1.0% 2.0% 3.0% 4.0% STD(R) E(R) Long A and Short B. Long A and Long B. Short A and Long B
F8000 Valuato of Facal ssets Sprg Semester 00 Dr. Isabel Tkatch ssstat Professor of Face Ivestmet Strateges Ledg vs. orrowg rsk-free asset) Ledg: a postve proporto s vested the rsk-free asset cash outflow
More information1036: Probability & Statistics
036: Probablty & Statstcs Lecture 9 Oe- ad Two-Sample Estmato Problems Prob. & Stat. Lecture09 - oe-/two-sample estmato cwlu@tws.ee.ctu.edu.tw 9- Statstcal Iferece Estmato to estmate the populato parameters
More informationSupplemental notes for topic 9: April 4, 6
Sta-30: Probablty Sprg 017 Supplemetal otes for topc 9: Aprl 4, 6 9.1 Polyomal equaltes Theorem (Jese. If φ s a covex fucto the φ(ex Eφ(x. Theorem (Beaymé-Chebyshev. For ay radom varable x, ɛ > 0 P( x
More informationScheduling of a Paper Mill Process Considering Environment and Cost
Schedulg of a Paper Mll Process Cosderg Evromet ad Cost M Park, Dogwoo Km, yog Km ad l Moo Departmet of Chemcal Egeerg, Yose Uversty, 34 Shchodog Seodaemooku, Seoul, 0-749, Korea Phoe: +8--363-9375 Emal:
More informationIEOR 130 Methods of Manufacturing Improvement Fall, 2017 Prof. Leachman Solutions to First Homework Assignment
IEOR 130 Methods of Maufacturg Improvemet Fall, 2017 Prof. Leachma Solutos to Frst Homework Assgmet 1. The scheduled output of a fab a partcular week was as follows: Product 1 1,000 uts Product 2 2,000
More informationPortfolio Selection and Optimization Based on DEA Cross Efficiency Evaluation
Iteratoal Joural Of Socal Scece a ucato eearch Vol. No.07 208 ISSN: 2637-6067 Abtract Portfolo Selecto a Optmzato Bae o DA Cro ffcecy valuato Fuwe L a Daa Zhou b Je Gao c School of coomc a Maagemet Chogqg
More information901 Notes: 16.doc Department of Economics Clemson University PRODUCTION THEORY 1
90 Notes: 6.doc Departmet of Ecoomcs Clemso Uversty PRODUCTION THEORY The eoclasscal theory of the frm s ot a theory about dustral orgazato but rather a theory about the relato betwee put ad output prces.
More informationChapter 4. More Interest Formulas
Chapter 4 More Iterest ormulas Uform Seres Compoud Iterest ormulas Why? May paymets are based o a uform paymet seres. e.g. automoble loas, house paymets, ad may other loas. 2 The Uform aymet Seres s 0
More informationA Test of Normality. Textbook Reference: Chapter 14.2 (eighth edition, pages 591 3; seventh edition, pages 624 6).
A Test of Normalty Textbook Referece: Chapter 4. (eghth edto, pages 59 ; seveth edto, pages 64 6). The calculato of p-values for hypothess testg typcally s based o the assumpto that the populato dstrbuto
More informationLecture 9 February 21
Math 239: Dscrete Mathematcs for the Lfe Sceces Sprg 2008 Lecture 9 February 21 Lecturer: Lor Pachter Scrbe/ Edtor: Sudeep Juvekar/ Alle Che 9.1 What s a Algmet? I ths lecture, we wll defe dfferet types
More informationSample Survey Design
Sample Survey Desg A Hypotetcal Exposure Scearo () Assume we kow te parameters of a worker s exposure dstrbuto of 8-our TWAs to a cemcal. As t appes, te worker as four dfferet types of days wt regard to
More informationForecasting the Movement of Share Market Price using Fuzzy Time Series
Iteratoal Joural of Fuzzy Mathematcs ad Systems. Volume 1, Number 1 (2011), pp. 73-79 Research Ida Publcatos http://www.rpublcato.com Forecastg the Movemet of Share Market Prce usg Fuzzy Tme Seres B.P.
More informationParametric Density Estimation: Maximum Likelihood Estimation
Parametric Desity stimatio: Maimum Likelihood stimatio C6 Today Itroductio to desity estimatio Maimum Likelihood stimatio Itroducto Bayesia Decisio Theory i previous lectures tells us how to desig a optimal
More informationTOPIC 7 ANALYSING WEIGHTED DATA
TOPIC 7 ANALYSING WEIGHTED DATA You do t have to eat the whole ox to kow that the meat s tough. Samuel Johso Itroducto dfferet aalyss for sample data Up utl ow, all of the aalyss techques have oly dealt
More informationRandom Variables. Discrete Random Variables. Example of a random variable. We will look at: Nitrous Oxide Example. Nitrous Oxide Example
Radom Varables Dscrete Radom Varables Dr. Tom Ilveto BUAD 8 Radom Varables varables that assume umercal values assocated wth radom outcomes from a expermet Radom varables ca be: Dscrete Cotuous We wll
More informationAMS Final Exam Spring 2018
AMS57.1 Fal Exam Sprg 18 Name: ID: Sgature: Istructo: Ths s a close book exam. You are allowed two pages 8x11 formula sheet (-sded. No cellphoe or calculator or computer or smart watch s allowed. Cheatg
More informationChapter 4. More Interest Formulas
Chapter 4 More Iterest ormulas Uform Seres Compoud Iterest ormulas Why? May paymets are based o a uform paymet seres. e.g. automoble loas, house paymets, ad may other loas. 2 The Uform aymet Seres s 0
More informationConsult the following resources to familiarize yourself with the issues involved in conducting surveys:
Cofdece Itervals Learg Objectves: After completo of ths module, the studet wll be able to costruct ad terpret cofdece tervals crtcally evaluate the outcomes of surveys terpret the marg of error the cotext
More informationOnline Encoding Algorithm for Infinite Set
Ole Ecodg Algorthm for Ifte Set Natthapo Puthog, Athast Surarers ELITE (Egeerg Laboratory Theoretcal Eumerable System) Departmet of Computer Egeerg Faculty of Egeerg, Chulalogor Uversty, Pathumwa, Bago,
More informationAn English Auction Protocol for Multi-Attribute Items
A Eglsh Aucto Protocol for Mult-Attrbute Items Esther Dav, Ra Azoulay-Schartz, Sart Kraus Bar-Ila Uversty, Departmet of Computer Scece, Ramat-Ga 5900, Israel {Saaoe, Schart, Sart}@macs.bu.ac.l Abstract.
More informationAn Efficient Estimator Improving the Searls Normal Mean Estimator for Known Coefficient of Variation
ISSN: 2454-2377, A Effcet Estmator Improvg the Searls Normal Mea Estmator for Kow Coeffcet of Varato Ashok Saha Departmet of Mathematcs & Statstcs, Faculty of Scece & Techology, St. Auguste Campus The
More informationLECTURE 5: Quadratic classifiers
LECURE 5: Quadratc classfers Bayes classfers for Normally dstrbuted classes Case : σ I Case : ( daoal) Case : ( o-daoal) Case : σ I Case 5: j eeral case Numercal example Lear ad quadratc classfers: coclusos
More informationIdentification of the De-synchronization, Synchronization and Forced Oscillation Phenomenon of a Nonlinear System
Ietfcato of the De-schrozato, Schrozato a Force Oscllato Pheomeo of a Nolear Sstem Marus-Costat O.S. Popescu, Osfor V. Olaru, Valeta E. Balas Abstract The pheomea of e-schrozato, schrozato, a force oscllato
More informationNaïve diversification, risk parity type I, minimum variance and risk parity type II (ERC) 1
Naïve dversfcato, rs arty tye I, mmum varace ad rs arty tye II (ERC) Harald Boger harald.boger@lve.de Frst Verso: Setember 9 th 05 Ths Verso: October 8 th 05 Cosder a mea-varace-otmzg vestor ad assume
More information= 1. UCLA STAT 13 Introduction to Statistical Methods for the Life and Health Sciences. Parameters and Statistics. Measures of Centrality
UCLA STAT Itroducto to Statstcal Methods for the Lfe ad Health Sceces Istructor: Ivo Dov, Asst. Prof. of Statstcs ad Neurolog Teachg Assstats: Brad Shaata & Tffa Head Uverst of Calfora, Los Ageles, Fall
More informationPortfolio Optimization. Application of the Markowitz Model Using Lagrange and Profitability Forecast
Epert Joural of Ecoomcs. Volume 6, Issue, pp. 6-34, 8 8 The Author. Publshed by Sprt Ivestfy. ISSN 359-774 Ecoomcs.EpertJourals.com Portfolo Optmzato. Applcato of the Markowtz Model Usg Lagrage ad Proftablty
More informationFeature Selection and Predicting CardioVascular Risk
Feature Selecto ad Predctg CardoVascular Rsk T.T.T.Nguye ad D.N. Davs, Computer Scece, Uversty of Hull. Itroducto No gold stadard ests for assessg the rsk of dvdual patets cardovascular medce. The medcal
More informationSolutions to Problems
Solutos to Problems ( Pt Pt + Ct) P5-. LG : Rate of retur: rt Pt Basc ($,000 $0,000 + $,500) a. Ivestmet X: Retur.50% $0,000 Ivestmet Y: Retur ($55,000 $55,000 + $6,800).36% $55,000 b. Ivestmet X should
More informationRSSI-based node localization algorithm for wireless sensor network
Avalable ole www.jocpr.com Joural of Chemcal a Pharmaceutcal Research, 14, 6(6):9-95 Research Artcle ISSN : 975-7384 CODEN(USA) : JCPRC5 -base oe localzato algorthm for wreless sesor etwork Wal Zhag* a
More informationMean-Semivariance Optimization: A Heuristic Approach
Mea-Semvarace Optmzato: A Heurstc Approach Javer Estrada IESE Busess School, Aveda Pearso 1, 08034 Barceloa, Spa Tel: +34 93 53 400, Fax: +34 93 53 4343, Emal: jestrada@ese.edu Abstract Academcs ad practtoers
More informationA New Method for Threshold Selection in Speech Enhancement by Wavelet Thresholding
011 Iteratoal Coerece o Computer Commucato ad Maagemet Proc.o CSIT vol.5 (011) (011) IACSIT Press, Sgapore A New Method or Threshold Selecto Speech hacemet b avelet Thresholdg Saeed Aat + * Assstat Proessor
More informationAPPENDIX M: NOTES ON MOMENTS
APPENDIX M: NOTES ON MOMENTS Every stats textbook covers the propertes of the mea ad varace great detal, but the hgher momets are ofte eglected. Ths s ufortuate, because they are ofte of mportat real-world
More informationMathematics 1307 Sample Placement Examination
Mathematcs 1307 Sample Placemet Examato 1. The two les descrbed the followg equatos tersect at a pot. What s the value of x+y at ths pot of tersecto? 5x y = 9 x 2y = 4 A) 1/6 B) 1/3 C) 0 D) 1/3 E) 1/6
More informationb. (6 pts) State the simple linear regression models for these two regressions: Y regressed on X, and Z regressed on X.
Mat 46 Exam Sprg 9 Mara Frazer Name SOLUTIONS Solve all problems, ad be careful ot to sped too muc tme o a partcular problem. All ecessary SAS fles are our usual folder (P:\data\mat\Frazer\Regresso). You
More informationPrediction Of Compressive Strength Of Concrete With Different Aggregate Binder Ratio Using ANN Model
Predcto Of Compressve Stregth Of Cocrete Wth Dfferet Aggregate Bder Rato Usg ANN Model Rama Moha Rao.P *, H.Sudarsaa Rao # * Assstat Professor (SG), Cetre for Dsaster Mtgato ad Maagemet, VIT Uversty, Vellore,
More informationOptions on the M Best of N Risky Assets 1
Optos o the M Best of Rsy Assets Tom Laamae Hels Uversty of Techology May 000 The author woul le to tha Matt Keloharu Teppo Martae Eva Lleblom a the partcpats of Hels chool of Ecoomcs Face Research emar
More informationThe Consumer Price Index for All Urban Consumers (Inflation Rate)
The Cosumer Prce Idex for All Urba Cosumers (Iflato Rate) Itroducto: The Cosumer Prce Idex (CPI) s the measure of the average prce chage of goods ad servces cosumed by Iraa households. Ths measure, as
More informationSorting. Data Structures LECTURE 4. Comparison-based sorting. Sorting algorithms. Quick-Sort. Example (1) Pivot
Data Structures, Sprg 004. Joskowcz Data Structures ECUE 4 Comparso-based sortg Why sortg? Formal aalyss of Quck-Sort Comparso sortg: lower boud Summary of comparso-sortg algorthms Sortg Defto Iput: A
More informationPORTFOLIO OPTIMIZATION IN THE FRAMEWORK MEAN VARIANCE -VAR
Lecturer Floret SERBAN, PhD Professor Vorca STEFANESCU, PhD Departmet of Mathematcs The Bucharest Academy of Ecoomc Studes Professor Massmlao FERRARA, PhD Departmet of Mathematcs Uversty of Reggo Calabra,
More informationNon-life insurance mathematics. Nils F. Haavardsson, University of Oslo and DNB Skadeforsikring
No-lfe surace mathematcs Nls F. Haavardsso, Uversty of Oslo ad DNB Skadeforskrg Repetto clam se The cocept No parametrc modellg Scale famles of dstrbutos Fttg a scale famly Shfted dstrbutos Skewess No
More informationInvestigating Signal Power Loss Prediction in A Metropolitan Island Using ADALINE and Multi-Layer Perceptron Back Propagation Networks
Ivestgatg Sgal Power Loss Predcto A Metropolta Islad Usg ADALINE ad Mult-Layer Perceptro Bac Propagato Networs Vrga Cha Ebhota 1*, Joseph Isaboa 2, Vrajay M. Srvastava 3 1, 2. 3 Departmet of Electroc Egeerg,
More informationValuation of Asian Option
Mälardales Uversty västerås 202-0-22 Mathematcs ad physcs departmet Project aalytcal face I Valuato of Asa Opto Q A 90402-T077 Jgjg Guo89003-T07 Cotet. Asa opto------------------------------------------------------------------3
More informationAlgorithm Analysis. x is a member of the set P x is not a member of the set P The null or empty set. Cardinality: the number of members
Algorthm Aalyss Mathematcal Prelmares: Sets ad Relatos: A set s a collecto of dstgushable members or elemets. The members are usually draw from some larger collecto called the base type. Each member of
More information? Economical statistics
Probablty calculato ad statstcs Probablty calculato Mathematcal statstcs Appled statstcs? Ecoomcal statstcs populato statstcs medcal statstcs etc. Example: blood type Dstrbuto A AB B Elemetary evets: A,
More informationMOMENTS EQUALITIES FOR NONNEGATIVE INTEGER-VALUED RANDOM VARIABLES
MOMENTS EQUALITIES FOR NONNEGATIVE INTEGER-VALUED RANDOM VARIABLES MOHAMED I RIFFI ASSOCIATE PROFESSOR OF MATHEMATICS DEPARTMENT OF MATHEMATICS ISLAMIC UNIVERSITY OF GAZA GAZA, PALESTINE Abstract. We preset
More informationPoverty indices. P(k;z; α ) = P(k;z; α ) /(z) α. If you wish to compute the FGT index of poverty, follow these steps:
Poverty dces DAD offers four possbltes for fxg the poverty le: - A determstc poverty le set by the user. 2- A poverty le equal to a proporto l of the mea. 3- A poverty le equal to a proporto m of a quatle
More informationCS 840 Fall 2018 Self-Organizing Binary Search Trees: Unit 3
S 840 Fall 2018 Self-Orgag ar Search Trees: Ut 3 The sae questos ca be asked bar search trees. Gve a sequece of access queres, what s the best wa to orgae the search tree [referece: ore Leserso, Rvest
More informationA Coverage Probability on the Parameters of the Log-Normal Distribution in the Presence of Left-Truncated and Right- Censored Survival Data ABSTRACT
Malaysa Joural of Mathematcal Sceces 9(1): 17-144 (015) MALAYSIAN JOURNAL OF MATHEMATICAL SCIENCES Joural homepage: http://espem.upm.edu.my/joural A Coverage Probablty o the Parameters of the Log-Normal
More information- Inferential: methods using sample results to infer conclusions about a larger pop n.
Chapter 6 Def : Statstcs: are commoly kow as umercal facts. s a feld of dscple or study. I ths class, statstcs s the scece of collectg, aalyzg, ad drawg coclusos from data. The methods help descrbe ad
More information. The firm makes different types of furniture. Let x ( x1,..., x n. If the firm produces nothing it rents out the entire space and so has a profit of
Joh Riley F Maimizatio with a sigle costrait F3 The Ecoomic approach - - shadow prices Suppose that a firm has a log term retal of uits of factory space The firm ca ret additioal space at a retal rate
More informationCOMPARISON OF APPROACHES TO TESTING EQUALITY OF EXPECTATIONS AMONG SAMPLES FROM POISSON AND NEGATIVE BINOMIAL DISTRIBUTION
ACTA UNIVERSITATIS AGRICULTURAE ET SILVICULTURAE MENDELIANAE BRUNENSIS Volume 66 0 Number 4, 08 https://do.org/0.8/actau08660405 COMPARISON OF APPROACHES TO TESTING EQUALITY OF EXPECTATIONS AMONG SAMPLES
More informationAn Entropy Method for Diversified Fuzzy Portfolio Selection
60 Iteratoal Joural of Fuzzy Systems, Vol. 4, No., March 0 A Etropy Method for Dversfed Fuzzy Portfolo Selecto Xaoxa Huag Abstract Ths paper proposes a etropy method for dversfed fuzzy portfolo selecto.
More informationSTOCK PRICE PREDICTION BY USING A COMBINATION OF NEURAL NETWORKS AND GENETIC ALGORITHMS
Arth Prabadh: A Joural of Ecoomcs ad Maagemet STOCK PRICE PREDICTION BY USING A COMBINATION OF NEURAL NETWORKS AND GENETIC ALGORITHMS MAHMOUD MOUSAVISHIRI*; FATEMEH SAEIDI** *Departmet of Maagemet, Ecoomcs
More informationFINANCIAL MATHEMATICS GRADE 11
FINANCIAL MATHEMATICS GRADE P Prcpal aout. Ths s the orgal aout borrowed or vested. A Accuulated aout. Ths s the total aout of oey pad after a perod of years. It cludes the orgal aout P plus the terest.
More informationManagement Science Letters
Maagemet Scece Letters (0) 355 36 Cotets lsts avalable at GrowgScece Maagemet Scece Letters homepage: www.growgscece.com/msl A tellget techcal aalyss usg eural etwork Reza Rae a Shapour Mohammad a ad Mohammad
More informationA Monte-Carlo Option-Pricing Algorithm for Log-Uniform Jump-Diffusion Model
A Mote-Carlo Opto-rcg Algorthm for Log-Uform Jump-Dffuso Model Zogwu Zhu ad Floyd B. Haso Abstract A reduced Europea call opto prcg formula by rsk-eutral valuato s gve. It s show that the Europea call
More informationThe Prediction Error of Bornhuetter-Ferguson
The Predcto Error of Borhuetter-Ferguso Thomas Mac Abstract: Together wth the Cha Ladder (CL method, the Borhuetter-Ferguso ( method s oe of the most popular clams reservg methods. Whereas a formula for
More informationFINANCIAL MATHEMATICS : GRADE 12
FINANCIAL MATHEMATICS : GRADE 12 Topcs: 1 Smple Iterest/decay 2 Compoud Iterest/decay 3 Covertg betwee omal ad effectve 4 Autes 4.1 Future Value 4.2 Preset Value 5 Skg Fuds 6 Loa Repaymets: 6.1 Repaymets
More informationMATHEMATICAL MODELLING OF RISK IN PORTFOLIO OPTIMIZATION WITH MEAN- EXTENDED GINI APPROACH
SCIREA Joural of Mathematcs http://www.screa.org/joural/mathematcs December 21, 2016 Volume 1, Issue 2, December 2016 MATHEMATICAL MODELLING OF RISK IN PORTFOLIO OPTIMIZATION WITH MEAN- EXTENDED GINI APPROACH
More informationStock price prediction
Международный Институт Экономики и Финансов ВЫПУСКНАЯ КВАЛИФИКАЦИОННАЯ РАБОТА на тему: Applcato of Collaboratve Flterg to Facal me Seres: Stock prce predcto Студент 4 курса Лобойко Арсений Александрович
More informationThe Statistics of Statistical Arbitrage
Volume 63 Number 5 007, CFA Isttute Robert Ferholz ad Cary Magure, Jr. Hedge fuds sometmes use mathematcal techques to capture the short-term volatlty of stocks ad perhaps other types of securtes. Ths
More informationMathematical Background and Algorithms
(Scherhet ud Zuverlässgket egebetteter Systeme) Fault Tree Aalyss Mathematcal Backgroud ad Algorthms Prof. Dr. Lggesmeyer, 0 Deftos of Terms Falure s ay behavor of a compoet or system that devates from
More informationBy choosing to view this document, you agree to all provisions of the copyright laws protecting it.
Copyrght 203 IEEE. Reprted, wth permsso, from Dgzhou Cao, Yu Su ad Huaru Guo, Optmzg Mateace Polces based o Dscrete Evet Smulato ad the OCBA Mechasm, 203 Relablty ad Mataablty Symposum, Jauary, 203. Ths
More informationInterest Rate Curve Estimation: a Financial Application for Support Vector Regression
Iterest Rate Curve Estmato: a Facal Applcato for Support Vector Regresso Adré d Almeda Motero Gavea Ivestmets Ro de Jaero Brazl amotero@gaveavest.com.br Frst draft: 07-ov-003 Ths draft: 07-u-004 Abstract
More informationPortfolio Optimization via Pair Copula-GARCH-EVT-CVaR Model
Avalable ole at www.scecedrect.com Systems Egeerg Proceda 2 (2011) 171 181 Portfolo Optmzato va Par Copula-GARCH-EVT-CVaR Model Lg Deg, Chaoqu Ma, Weyu Yag * Hua Uversty, Hua, Chagsha 410082, PR Cha Abstract
More informationThe Application of Asset Pricing to Portfolio Management
Clemso Ecoomcs The Applcato of Asset Prcg to Portfolo Maagemet The Nature of the Problem Portfolo maagers have two basc problems. Frst they must determe whch assets to hold a portfolo, ad secod, they must
More informationApplication of Portfolio Theory to Support Resource Allocation Decisions for Biosecurity
Applcato of Portfolo Theory to Support Resource Allocato Decsos for Bosecurty Paul Mwebaze Ecoomst 11 September 2013 CES/BIOSECURITY FLAGSHIP Presetato outle The resource allocato problem What ca ecoomcs
More informationPrediction Error of the Future Claims Component of Premium Liabilities under the Loss Ratio Approach. International Regulatory Changes
Predcto rror o the Future lams ompoet o Premum Labltes uder the Loss Rato Approach (accepted to be publshed ace) AS Aual Meetg November 8 00 Jacke L PhD FIAA Nayag Busess School Nayag Techologcal Uversty
More informationDetermining Optimal Production Program with Fuzzy Multiple Criteria Programming Method
Proceedgs o the Iteratoal MultCoerece o Egeers ad Computer Scetsts 9 Vol II IMECS 9 March - 9 Hog og Determg Optmal Producto Program wth Fuzzy Multple Crtera Programmg Method Tujo Perć Member IAENG Zora
More informationNonlinear Modeling of European Football Scores Using Support Vector Machines
Nolear Modelg of Europea Football Scores Usg Support Vector Maches Raphael Ncholas Markellos To cte ths verso: Raphael Ncholas Markellos. Nolear Modelg of Europea Football Scores Usg Support Vector Maches.
More informationAllocation of Optimal Distributed Generation using GA for Minimum System Losses
Ffteeth Natoal ower Systems Coferece (NSC), IIT Bombay, December 2008 Allocato of Optmal Dstrbuted Geerato usg GA for Mmum System Losses T.N. Shukla, S.. Sgh ad K.B. Nak Abstract Dstrbuted Geerato () represets
More informationCopula based simulation procedures for pricing. basket Credit Derivatives
Copula based smulato procedures for prcg basket Credt Dervatves Fath Abd* Faculty of Busess ad Ecoomcs, Uversty of Sfax, UR: MODESFI, Sfax, Tusa Nader Nafar Isttute of the Hgher Busess Studes, Uversty
More informationThe Measurement and Control of Chinese Administrative Expenses: Perspective into Administrative Expenses
Joural of Poltcs ad Law Jue, 9 The Measuremet ad Cotrol of Chese Admstratve Epeses: Perspectve to Admstratve Epeses Xagzhou He Zhejag Uversty Hagzhou 38, Cha E-mal: hez5@6.com Natoal Natural Scece Foudato
More informationSTATIC GAMES OF INCOMPLETE INFORMATION
ECON 10/410 Decsos, Markets ad Icetves Lecture otes.11.05 Nls-Herk vo der Fehr SAIC GAMES OF INCOMPLEE INFORMAION Itroducto Complete formato: payoff fuctos are commo kowledge Icomplete formato: at least
More informationDynamic Economic Load Dispatch of Electric Power System Using Direct Method
Iteratoal Joural of Appled Egeerg Research ISSN 0973-456 Volume 13, Number 6 (018) pp. 34-347 Research Ida ublcatos. http://www.rpublcato.com Dyamc Ecoomc oad Dspatch of Electrc ower System Usg Drect Method
More informationA polyphase sequences with low autocorrelations
oural o Physcs: Coerece Seres PAPER OPE ACCESS A polyphase sequeces wth low autocorrelatos To cte ths artcle: A Leuh 07. Phys.: Co. Ser. 859 00 Vew the artcle ole or updates ad ehacemets. Related cotet
More informationUncertain Supply Chain Management
Ucerta Supply Cha Maagemet (0) 07 Cotets lsts avalable at GrowgScece Ucerta Supply Cha Maagemet homepage: www.growgscece.com/uscm A methodology for outsourcg resources reverse logstcs usg fuzzy TOPSIS
More informationMay 2005 Exam Solutions
May 005 Exam Soluto 1 E Chapter 6, Level Autes The preset value of a auty-mmedate s: a s (1 ) v s By specto, the expresso above s ot equal to the expresso Choce E. Soluto C Chapter 1, Skg Fud The terest
More informationANALYSING COMMON STOCKS PERFORMANCE FROM THE OPTIMAL EX-POST PORTFOLIO WEIGHTS. This version: October 2014
ANALYSING COON STOCKS PERFORANCE FRO THE OPTIAL EX-POST PORTFOLIO WEIGHTS d Face Forum Zaragoza 0 th November 014 Ths verso: October 014 ara-teresa Bosch-Bada Vstg Researcher Uverstat de Groa Departmet
More informationMACHINE LEARNING OF EXPERT DECISION OR SYSTEM BEHAVIOUR. Peter Otto
MACHINE LEARNING OF EXPERT DECISION OR SYSTEM BEHAVIOUR Peter Otto Isttut für Automatserugs- ud Systemtechk Techsche Uverstät Ilmeau Gustav-Krchhoff-Straße D-98693 Ilmeau, Germay Emal: peter.otto@tu-lmeau.de
More informationProceedings: National Workshop on IT Services and Applications (WITSA2003) Feb 27-28, 2003
Proceedgs: Natoal Workshop o IT ervces ad Applcatos (WITA2003) Feb 27-28, 2003 kew Estmato of Bary ocumet Images Usg tatc ad yamc Thresholds Useful for ocumet Image osacg *P.hvakumara, G. Hematha Kumar,.
More informationChapter Four 1/15/2018. Learning Objectives. The Meaning of Interest Rates Future Value, Present Value, and Interest Rates Chapter 4, Part 1.
Chapter Four The Meaig of Iterest Rates Future Value, Preset Value, ad Iterest Rates Chapter 4, Part 1 Preview Develop uderstadig of exactly what the phrase iterest rates meas. I this chapter, we see that
More informationATutorialonParticleFilteringandSmoothing: Fifteen years later
ATutoraloPartcleFltergadSmoothg: Fftee years later Araud Doucet The Isttute of Statstcal Mathematcs, 4-6-7 Mam-Azabu, Mato-ku, Tokyo 06-8569, Japa Emal: Araud@smacjp Adam M Johase Departmet of Statstcs,
More information