A GENERALIZED ADDITIVE LOGIT MODEL OF BRAND CHOICE * SUNIL K SAPRA California State University Los Angeles, United States

Size: px
Start display at page:

Download "A GENERALIZED ADDITIVE LOGIT MODEL OF BRAND CHOICE * SUNIL K SAPRA California State University Los Angeles, United States"

Transcription

1 Journal of Contemorary Issues n Busness Research Volume, Issue No. 3, 22 Journal of Contemorary Issues n Busness Research ISSN (Onlne), 22, Vol., No. 3, Coyrght of the Academc Journals JCIBR All rghts reserved. A GENERALIZED ADDITIVE LOGIT MODEL OF BRAND CHOICE * SUNIL K SAPRA Calforna State Unversty Los Angeles, Unted States ABSTRACT The aer resents an econometrc alcaton of generalzed addtve Logt regresson models (GALRMs) for brand choce. Our sem-arametrc models are flexble and robust extensons of the Logt model. The GALRMs are ft to bnary resonse data by maxmzng a enalzed log lkelhood or a enalzed log artal-lkelhood. The GAMs allow us to buld a regresson surface as a sum of lower-dmensonal nonarametrc terms crcumventng the curse of dmensonalty: the slow convergence of an estmator to the true value n hgh dmensons. Four GALRMs are comared wth a Logt model for brand choce and the best model s selected usng varous model selecton crtera. Keywords: Logt Model; Generalzed Addtve Model; Interactons; Backfttng Algorthm; Penalzed Regresson Slnes. INTRODUCTION Logt models are ervasve n busness and economcs. The most common alcatons of Logt models have been to the analyss of brand choce data n marketng (Baltas 997 and Guadagn and Lttle 983) and transortaton choce data n economcs (Greene 28 and Mansk and McFadden 983). Logt models belong to the class of generalzed lnear models, whch relax the assumton that the resonse s normally dstrbuted by allowng t to follow any dstrbuton from the exonental famly (for examle, normal, Posson, bnomal, gamma etc.). Inference for GLMs s based on lkelhood theory. McCullagh and Nelder (989) rovde an authortatve account of GLMs and Cameron and Trved (25) and Greene (28) rovde econometrc alcatons. In recent years, sem-arametrc extensons of lnear regressons have been emloyed n economcs and busness. Nevertheless, there have been relatvely few alcatons of semarametrc extensons of generalzed lnear models. Ths aer s an attemt n ths drecton. We resent an econometrc alcaton of a generalzed addtve model (GAM) to bnary choce data, whch s a sem-arametrc extenson of GLMs. A GAM s a sem-arametrc GLM n whch art of the lnear redctor s secfed n terms of a sum of unknown smooth functons of exlanatory varables. Generalzed addtve models (GAMs) are a owerful generalzaton of lnear, logstc, and Posson regresson models. GAMs are very flexble, and can rovde an excellent ft n the resence of nonlnear relatonshs. GAMs and GLMs can be aled n smlar stuatons, but they serve dfferent analytc uroses. GLMs emhasze estmaton and nference for the arameters of the model, whle GAMs focus on exlorng * The vews or onons exressed n ths manuscrt are those of the author(s) and do not necessarly reflect the oston, vews or onons of the edtor(s), the edtoral board or the ublsher. Corresondng author Emal: ssara@exchange.calstatela.edu 69

2 Journal of Contemorary Issues n Busness Research Volume, Issue No. 3, 22 data nonarametrcally. The GAM aroach offers more flexblty n model form than the GLM aroach does. These models estmate an addtve aroxmaton to the multvarate lnk functon. The key beneft of ths aroach s that each of the ndvdual addtve terms s estmated usng a unvarate smoother nstead of a multvarate smoother for a hghdmensonal non-arametrc term crcumventng the curse of dmensonalty: the slow convergence of an estmator to the true value n hgh dmensons. The GAM formulaton of the Logt regresson model allows us to buld a regresson surface as a sum of lowerdmensonal nonarametrc terms. The two man aroaches to estmaton of GAMs are backfttng wth local scorng algorthm (Haste and Tbshran (99)) and enalzed regresson slnes (Wood (26)). Backfttng has the advantage that t can be used wth any scatterlot smoother whle the latter method has the advantage that estmaton of the smoothng arameter usng generalzed cross valdaton s ntegrated nto estmaton. Ths aer resents an econometrc alcaton of the GAM extenson of the Logt model and demonstrates that t can overcome a serous weakness of the Logt model n falng to dentfy the nonlneartes n the lnk functon. The aer s organzed as follows. Secton 2 ntroduces the generalzed addtve Logt regresson model (GALRM). Secton 3 resents the enalzed regresson method for the estmaton of GALRMs. Secton 4 resents an econometrc alcaton of GALRM to Coke data on choce between Coke and Pes. Secton 5 rovdes some concludng remarks. GENERALIZED ADDITIVE MODELS Generalzed addtve models are nonarametrc generalzed lnear models. Generalzed addtve models extend tradtonal lnear models n another way, namely by allowng for a lnk between the nonlnear redctor f(x,...,x) and the exected value of y. Ths amounts to allowng for an alternatve dstrbuton for the underlyng random varaton besdes ust the normal dstrbuton. Whle Gaussan models can be used n many statstcal alcatons, for several tyes of roblems they are not arorate. For examle, the normal dstrbuton may not be adequate for modelng dscrete resonses such as counts, or bounded resonses such as roortons. Generalzed addtve models consst of a random comonent, an addtve comonent, and a lnk functon relatng these two comonents. The resonse y, the random comonent, s assumed to have a densty n the exonental famly. f Y y b( ) ( y;, ) ex{ c( y. ) a( ) Where s called the natural arameter and Φ s the scale arameter. The normal, bnomal, and Posson dstrbutons are all n ths famly. Unlke the generalzed lnear models, the mean μ = E(y x, x2,, x) s not lnked to the lnear redctor xβ, but to the nonlnear nonarametrc redctor. η = g( ) ( ), s x Where s( ),..., s( ) are smooth nonarametrc functons defnes the addtve comonent. Fnally, the relatonsh between the mean μ of the resonse varable and η s defned by a lnk functon g(μ) = η. The most commonly used lnk functon s the canoncal lnk, for whch η = θ. A combnaton of backfttng and local scorng algorthms are used n the actual fttng of the model. In order to ft GAMs to the data, followng Wood (26), we use bass exansons of smooth functons and enalzed lkelhood maxmzaton for model estmaton 7

3 Journal of Contemorary Issues n Busness Research Volume, Issue No. 3, 22 n whch wggly models are enalzed more heavly than smooth models n a controllable manner, and degree of smoothness s chosen based on cross valdaton, or AIC or Mallows crteron. The generalzed addtve models are ft to count data or bnary resonse data by maxmzng a enalzed log lkelhood or a enalzed log artal-lkelhood. To maxmze t, the backfttng rocedure s used n conuncton wth a maxmum lkelhood or maxmum artal lkelhood algorthm (Haste and Tbshran (99) and Wood (26)). The Newton- Rahson method for maxmzng log-lkelhoods n these models can be resented n an IRLS (teratvely reweghted least squares) form. It nvolves a reeated weghted lnear regresson of a constructed resonse varable on the covarates: each regresson yelds a new value of the arameter estmates whch gve a new constructed varable, and the rocess s terated. In the generalzed addtve model, the weghted lnear regresson s smly relaced by a weghted backfttng algorthm (Haste and Tbshran (99)). ESTIMATION OF GENERALIZED ADDITIVE LOGIT REGRESSION MODEL USING PENALIZED REGRESSION METHOD Algorthm for Penalzed Iteratvely Re-weghted Least Squares (PIRLS) (Wood (26) The GALRMs are ft to bnary resonse data by maxmzng a enalzed log lkelhood or a enalzed log artal-lkelhood. The followng algorthm s used to mlement these methods. The R-ackage mgcv (Wood (22) ) was used for comutatons. Ste : Intalze ˆ g(/ N N y ), s,,2,...,. Ste 2 :Construct an adusted deendent varable z as z ( y )( / g( ) s ( x Ste 3:Comuteweghts w ( / ) ( V ), ) ) and g where V sthe varance of y at, Ste 4: Penalzed Slne Re gresson Mnmze smoothng arameter.comutes of 2 ( ). the matrx of data on bass functons used to reresent the regresson functon, W s a dagonal matrx wth - th dagonal element matrx of known coeffcentsn the enalty functon ' S and s a s,, and. W ( z X ) ' S wth resect to, where X s convergenc e s obtaned. Ste 5:Reeat stes2-4 relacng,, and, the second stage estmates w, S s a by untl the dfference between twosuccessve values of s less than a small resecfed number and The Generalzed Addtve Logt Model The generalzed addtve Logt model assumes that 7

4 Journal of Contemorary Issues n Busness Research Volume, Issue No. 3, 22 g( ) logt( ) ln where E( y x ) ex{ s ( x ), s ( x )}/ ex{ s ( x )}. The adusted deendent varable z and the weghts w used n the algorthm above are z ( y ) / ( ), w ( ) where g ( ), s ( x ) The functons s, s 2,, s are estmated by an algorthm lke the one descrbed earler. AN EMPIRICAL APPLICATION OF GAM LOGIT MODEL TO BRAND CHOICE GAM Logt Aled to Coke Data The dataset s from ERIM ublc data base, James M. Klts Center, Unversty of Chcago Booth School of Busness and conssts of data on 4 ndvduals who urchased Coke or Pes. Data Descrton The deendent varable COKE s defned as follows: COKE = f Coke s chosen, = f Pes s chosen PR_PEPSI = Prce of 2 lter bottle of Pes PR_COKE = Prce of 2 lter bottle of Coke DISP_PEPSI = f Pes s dslayed at tme of urchase, otherwse = DISP_COKE = f Coke s dslayed at tme of urchase, otherwse = PRATIO = Prce of Coke relatve to rce of Pes TABLE Summary Statstcs Varable Obs. Mean Std. Dev. Mn Max COKE PR_PEPSI PR_COKE DISP_PEPSI DISP_COKE PRATIO Source. ERIM ublc data base, James M. Klts Center, Unversty of Chcago Booth School of Busness Models The followng models were ft to the data. Model s a Logt model. Gven the nonlnearty of the Logt lnk functon n PRATIO as dslayed n the artal resdual lot of PRATIO n Fg., Model 2, a Generalzed Addtve Logt model ntroduces a nonarametrc smooth term s(pratio). The ossblty of nteracton between PRATIO and DISP_PEPSI and between PRATIO and DISP_COKE was nvestgated n models 3 and 4 resectvely by ntroducng a nonarametrc nteracton term n each model. Fnally, Model 5 s a GLM Logt 72

5 Journal of Contemorary Issues n Busness Research Volume, Issue No. 3, 22 model, whch ntroduces arametrc nteractons between PRATIO and DISP_PEPSI as well as between PRATIO and DISP_COKE. FIGURE Partal Resdual Plot of V6 Model : Logt Re gresson Model g( ) DISP _ PEPSI DISP _ COKE PRATIO Model 2 : Generalzed Addtve Logt Re gresson Models g( ) DISP _ PEPSI DISP _ COKE s( PRATIO) 4 5 Model 3 : Generalzed Addtve Logt Re gresson Models wth a Nonarametrc Interacto nterm g( ) DISP _ COKE s( PRATIO * DISP _ PEPSI ) 5 Model 4 : Generalzed Addtve Logt Re gresson Models wth a Nonarametrc Interacto nterm g( ) DISP _ PEPSI s( PRATIO * DISP _ COKE) 4 Model 5 : Logt Re gresson Model wth Interacto ns between Prato and DISP _ PEPSI and between Prato and DISP _ COKE g( ) DISP _ PEPSI DISP _ COKE PRATIO PRATIO * DISP _ PEPSI PRATIO * DISP _ COKE

6 Journal of Contemorary Issues n Busness Research Volume, Issue No. 3, 22 TABLE 2 Model : GLM Logt: Coke Data Varable Coeffcent Estmate Std. Error t-rato -value Intercet x -9*** DISP_PEPSI x -5*** DISP_COKE * PRATIO x -*** Sgnf. codes: ***. **. *.5.. (Dserson arameter for bnomal famly taken to be ) Null devance: on 39 degrees of freedom Resdual devance: 48.9 on 36 degrees of freedom AIC = Number of Fsher Scorng teratons: 3 TABLE 3 Model 2: GAM Logt: Coke Data Varable Coeffcent Estmate Std. Error t-rato -value Intercet DISP_PEPSI *** DISP_COKE * Aroxmate sgnfcance of smooth terms: Varable Estmated df Refned df Ch-squared -value s(pratio) x - ** Sgnf. codes: ***. **. *.5.. R-sq.(ad) =.35 Devance exlaned =.% UBRE score =.2483 Scale est. = n = 4 AIC = Null Devance: on 39 degrees of freedom Resdual Devance: on 33 degrees of freedom Number of Local Scorng Iteratons: 8 74

7 Journal of Contemorary Issues n Busness Research Volume, Issue No. 3, 22 TABLE 4 Model 3: GAM Logt Model wth a Smooth Nonarametrc Functon and Interacton between V4 and V6: Coke Data Varable Coeffcent Estmate Std. Error t-rato -value Intercet DISP_PEPSI DISP_COKE ** Aroxmate sgnfcance of smooth terms: Varable Estmated df Refned df Ch-squared -value s(pratio) x -2 *** s(disp_pepsi*pratio) *** Sgnf. codes: ***. **. *.5.. R-sq.(ad) =.42 Devance exlaned =.5% UBRE score = Scale est. = n = 4 AIC = Null Devance: on 39 degrees of freedom Resdual Devance: on 29 degrees of freedom Number of Local Scorng Iteratons: 2 TABLE 5 Model 4: GAM Logt Model wth a Smooth Nonarametrc Functon and Interacton between V5 and V6: Coke Data Varable Coeffcent Estmate Std. Error t-rato -value Intercet DISP_PEPSI * DISP_COKE Aroxmate sgnfcance of smooth terms: Varable Estmated df Refned df Ch-squared -value s(pratio) x -6 *** s(disp_coke*pratio) Sgnf. codes: ***. **. *.5.. R-sq.(ad) =.42 Devance exlaned = 2% UBRE score =.2375 Scale est. = n = 4 AIC = Null Devance: on 39 degrees of freedom Resdual Devance: on 29 degrees of freedom Number of Local Scorng Iteratons: 9 75

8 Journal of Contemorary Issues n Busness Research Volume, Issue No. 3, 22 TABLE 6 Model 5: GLM Logt wth Interacton: Coke Data Varable Coeffcent Estmate Std. Error t-rato -value Intercet x -8*** DISP_PEPSI x -5*** DISP_COKE * PRATIO x -9*** DISP_PEPSI*PRATIO ** DISP_COKE*PRATIO Sgnf. codes: ***. **. *.5.. (Dserson arameter for bnomal famly taken to be ) Null devance: on 39 degrees of freedom Resdual devance: 4.7 on 34 degrees of freedom AIC: 42.7 Number of Fsher Scorng teratons: 4 Comarng the Models The estmaton and results are resented n tables 2 through 6. A comarson of models usng the AIC s resented n Table 7. TABLE 7 Models and the AICs MODEL AIC Model 3, a modfed GAM, whch allows for nonarametrc nteracton between a dscrete varable DISP_PEPSI and a contnuous varable PRATIO, has the lowest AIC and UBRE score among the fve models studed and s therefore the best model. The Logt regresson model (Model ) has the hghest AIC. Ths s not surrsng snce the Logt model msses the nonlnearty n PRATIO n the lnk functon. Model 3 also has the lowest devance of on 29 degrees of freedom, whle Model has the hghest devance of 48.9 on 36 degrees of freedom. The statstcal sgnfcance of PRATIO dffers markedly between the Logt regresson model and the varous GAM Logt models emloyed here. The sgns of the coeffcent estmates are all exected n all of the models but Model 4. For nstance, the sgn of DISP_PEPSI s negatve and the sgn of DISP_COKE s ostve across models through 3 and model 5 ndcatng that the odds of choosng Coke over Pes ncrease f Pes s dslayed at the tme of urchase and decrease f Coke s dslayed at the tme of urchase. The analyss of devance n Tables 2, 3, and 4 ndcates sgnfcant nonlnear contrbuton from the varable PRATIO. The nonarametrc nteracton term n Model 3 for nonlnear nteracton between DISP_PEPSI and PRATIO turns out to be hghly sgnfcant ndcatng another nonlnearty n the lnk functon mssed by the Logt model. The arametrc lnear nteracton term for nteracton between DISP_PEPSI and PRATIO n Model 5 also turns out to be hghly sgnfcant. Nevertheless, Table 5 ndcates a weak nteracton between nonlnear effects of PRATIO and the varable DISP_COKE. The hgh degree of nonlnearty n PRATIO s also seen n the artal resdual smoothng lot of PRATIO n Fgure. The dotted curves around the sold curve reresent +-2 standard errors 76

9 Journal of Contemorary Issues n Busness Research Volume, Issue No. 3, 22 around the sold curve. The only surrsng result s the negatve sgn of the varable DISP_COKE n Table 4. CONCLUSIONS The aer has studed four generalzed addtve Logt regresson models (GALRMs) as alternatves to the Logt regresson model. In all of the emrcal alcatons, each of the GALRMs rovdes a much better ft than the Logt regresson model as reflected n lower AICs and lower devances. The econometrc technques used n the aer are wdely alcable to the analyss of count, bnary resonse and duraton tyes of data occurrng frequently n economcs and busness. Nevertheless, the GALRMs are not wthout drawbacks. The comutatonal algorthms are comlex and nterretatons can be dffcult. These models are useful manly when smle models for the lnear redctor rovde an nadequate ft for the data. REFERENCES Baltas, G. (997). Determnants of store brand choce: a behavoral analyss. Journal of Product & Brand Management, 6 (5), Cameron, A. C., and Trved, P. (998). Mcroeconometrcs. Cambrdge Unversty Press. Greene, W. H. (28). Econometrc Analyss, Pearson/Prentce Hall, New York. Guadagn, P. M., and Lttle, John, D.C. (983). A Logt Model of Brand Choce Calbrated on Scanner Data. Marketng Scence Summer Vol. 2 (3), Haste, T., & Tbshran, R. (99). Generalzed Addtve Models. Chaman and Hall, London. McCullagh, P., and Nelder, J. (989). Generalzed Lnear Models. Chaman and Hall, London. Mansk, C., and Danel McFadden (98). Structural Analyss of Dscrete Data and Econometrc Alcatons. MIT Press, Cambrdge. Wood, S. N. (22), R-ackage mgcv. Download From htt://cran.rroect.org/web/ackages/mgcv/mgcv.df Wood, S. N. (26). Generalzed Addtve Models: an ntroducton wth R. CRC, Boca Raton. 77

OMICS Group. Contact us at:

OMICS Group. Contact us at: OMICS Group OMICS Group Internatonal through ts Open Access Intatve s commtted to make genune and relable contrbutons to the scentfc communty. OMICS Group hosts over 4 leadng-edge peer revewed Open Access

More information

NAG Fortran Library Chapter Introduction. G10 Smoothing in Statistics

NAG Fortran Library Chapter Introduction. G10 Smoothing in Statistics Introducton G10 NAG Fortran Lbrary Chapter Introducton G10 Smoothng n Statstcs Contents 1 Scope of the Chapter... 2 2 Background to the Problems... 2 2.1 Smoothng Methods... 2 2.2 Smoothng Splnes and Regresson

More information

Speed of price adjustment with price conjectures

Speed of price adjustment with price conjectures Seed of rce adustment wh rce conectures Mchael Olve Macquare Unversy, Sydney, Australa Emal: molve@efs.mq.edu.au Abstract We derve a measure of frm seed of rce adustment that s drectly nversely related

More information

Assessing maximally allowed concentration; application to the regulation of PNOS

Assessing maximally allowed concentration; application to the regulation of PNOS Assessng maxmally allowed concentraton; alcaton to the regulaton of PNOS G. SALANTI, K. ULM Insttut für Medznsche Statstk und Edemologe der TU München Klnkum rechts der Isar Ismannger Str. 22 8675 Muenchen

More information

CS 534: Computer Vision Model Fitting

CS 534: Computer Vision Model Fitting CS 534: Computer Vson Model Fttng Sprng 004 Ahmed Elgammal Dept of Computer Scence CS 534 Model Fttng - 1 Outlnes Model fttng s mportant Least-squares fttng Maxmum lkelhood estmaton MAP estmaton Robust

More information

Smoothing Spline ANOVA for variable screening

Smoothing Spline ANOVA for variable screening Smoothng Splne ANOVA for varable screenng a useful tool for metamodels tranng and mult-objectve optmzaton L. Rcco, E. Rgon, A. Turco Outlne RSM Introducton Possble couplng Test case MOO MOO wth Game Theory

More information

Exercises (Part 4) Introduction to R UCLA/CCPR. John Fox, February 2005

Exercises (Part 4) Introduction to R UCLA/CCPR. John Fox, February 2005 Exercses (Part 4) Introducton to R UCLA/CCPR John Fox, February 2005 1. A challengng problem: Iterated weghted least squares (IWLS) s a standard method of fttng generalzed lnear models to data. As descrbed

More information

Index Terms-Software effort estimation, principle component analysis, datasets, neural networks, and radial basis functions.

Index Terms-Software effort estimation, principle component analysis, datasets, neural networks, and radial basis functions. ISO 9001:2008 Certfed Internatonal Journal of Engneerng and Innovatve Technology (IJEIT The Effect of Dmensonalty Reducton on the Performance of Software Cost Estmaton Models Ryadh A.K. Mehd College of

More information

X- Chart Using ANOM Approach

X- Chart Using ANOM Approach ISSN 1684-8403 Journal of Statstcs Volume 17, 010, pp. 3-3 Abstract X- Chart Usng ANOM Approach Gullapall Chakravarth 1 and Chaluvad Venkateswara Rao Control lmts for ndvdual measurements (X) chart are

More information

Lecture 4: Principal components

Lecture 4: Principal components /3/6 Lecture 4: Prncpal components 3..6 Multvarate lnear regresson MLR s optmal for the estmaton data...but poor for handlng collnear data Covarance matrx s not nvertble (large condton number) Robustness

More information

EXST7034 Regression Techniques Geaghan Logistic regression Diagnostics Page 1

EXST7034 Regression Techniques Geaghan Logistic regression Diagnostics Page 1 Logstc regresson Dagnostcs Page 1 1 **********************************************************; 2 *** Logstc Regresson - Dsease outbreak example ***; 3 *** NKNW table 14.3 (Appendx C3) ***; 4 *** Study

More information

Empirical Distributions of Parameter Estimates. in Binary Logistic Regression Using Bootstrap

Empirical Distributions of Parameter Estimates. in Binary Logistic Regression Using Bootstrap Int. Journal of Math. Analyss, Vol. 8, 4, no. 5, 7-7 HIKARI Ltd, www.m-hkar.com http://dx.do.org/.988/jma.4.494 Emprcal Dstrbutons of Parameter Estmates n Bnary Logstc Regresson Usng Bootstrap Anwar Ftranto*

More information

Region Segmentation Readings: Chapter 10: 10.1 Additional Materials Provided

Region Segmentation Readings: Chapter 10: 10.1 Additional Materials Provided Regon Segmentaton Readngs: hater 10: 10.1 Addtonal Materals Provded K-means lusterng tet EM lusterng aer Grah Parttonng tet Mean-Shft lusterng aer 1 Image Segmentaton Image segmentaton s the oeraton of

More information

Subspace clustering. Clustering. Fundamental to all clustering techniques is the choice of distance measure between data points;

Subspace clustering. Clustering. Fundamental to all clustering techniques is the choice of distance measure between data points; Subspace clusterng Clusterng Fundamental to all clusterng technques s the choce of dstance measure between data ponts; D q ( ) ( ) 2 x x = x x, j k = 1 k jk Squared Eucldean dstance Assumpton: All features

More information

y and the total sum of

y and the total sum of Lnear regresson Testng for non-lnearty In analytcal chemstry, lnear regresson s commonly used n the constructon of calbraton functons requred for analytcal technques such as gas chromatography, atomc absorpton

More information

Trends in Computer Science and Information Technology

Trends in Computer Science and Information Technology Engneerng Grou Trends n Comuter Scence and Informaton Technology DOI htt://dx.do.org/0.735/tcst.000005 DOI CC By Eren Bas *, Erol Egroglu and Vedde Rezan Uslu Deartment of Statstcs, Faculty of Arts and

More information

S1 Note. Basis functions.

S1 Note. Basis functions. S1 Note. Bass functons. Contents Types of bass functons...1 The Fourer bass...2 B-splne bass...3 Power and type I error rates wth dfferent numbers of bass functons...4 Table S1. Smulaton results of type

More information

A Semi-parametric Regression Model to Estimate Variability of NO 2

A Semi-parametric Regression Model to Estimate Variability of NO 2 Envronment and Polluton; Vol. 2, No. 1; 2013 ISSN 1927-0909 E-ISSN 1927-0917 Publshed by Canadan Center of Scence and Educaton A Sem-parametrc Regresson Model to Estmate Varablty of NO 2 Meczysław Szyszkowcz

More information

Feature Reduction and Selection

Feature Reduction and Selection Feature Reducton and Selecton Dr. Shuang LIANG School of Software Engneerng TongJ Unversty Fall, 2012 Today s Topcs Introducton Problems of Dmensonalty Feature Reducton Statstc methods Prncpal Components

More information

Support Vector Machines

Support Vector Machines /9/207 MIST.6060 Busness Intellgence and Data Mnng What are Support Vector Machnes? Support Vector Machnes Support Vector Machnes (SVMs) are supervsed learnng technques that analyze data and recognze patterns.

More information

Phd Program in Transportation. Transport Demand Modeling. Session 7

Phd Program in Transportation. Transport Demand Modeling. Session 7 Phd Program n Transportaton Transport Demand Modelng Lus Martínez (based on the Lessons of Anabela Rbero TDM2010) Sesson 7 Generalzed Lnear Models Phd n Transportaton / Transport Demand Modellng 1/48 Outlne

More information

Biostatistics 615/815

Biostatistics 615/815 The E-M Algorthm Bostatstcs 615/815 Lecture 17 Last Lecture: The Smplex Method General method for optmzaton Makes few assumptons about functon Crawls towards mnmum Some recommendatons Multple startng ponts

More information

The Man-hour Estimation Models & Its Comparison of Interim Products Assembly for Shipbuilding

The Man-hour Estimation Models & Its Comparison of Interim Products Assembly for Shipbuilding Internatonal Journal of Operatons Research Internatonal Journal of Operatons Research Vol., No., 9 4 (005) The Man-hour Estmaton Models & Its Comparson of Interm Products Assembly for Shpbuldng Bn Lu and

More information

Learning the Kernel Parameters in Kernel Minimum Distance Classifier

Learning the Kernel Parameters in Kernel Minimum Distance Classifier Learnng the Kernel Parameters n Kernel Mnmum Dstance Classfer Daoqang Zhang 1,, Songcan Chen and Zh-Hua Zhou 1* 1 Natonal Laboratory for Novel Software Technology Nanjng Unversty, Nanjng 193, Chna Department

More information

An Iterative Solution Approach to Process Plant Layout using Mixed Integer Optimisation

An Iterative Solution Approach to Process Plant Layout using Mixed Integer Optimisation 17 th European Symposum on Computer Aded Process Engneerng ESCAPE17 V. Plesu and P.S. Agach (Edtors) 2007 Elsever B.V. All rghts reserved. 1 An Iteratve Soluton Approach to Process Plant Layout usng Mxed

More information

Classifier Selection Based on Data Complexity Measures *

Classifier Selection Based on Data Complexity Measures * Classfer Selecton Based on Data Complexty Measures * Edth Hernández-Reyes, J.A. Carrasco-Ochoa, and J.Fco. Martínez-Trndad Natonal Insttute for Astrophyscs, Optcs and Electroncs, Lus Enrque Erro No.1 Sta.

More information

Three supervised learning methods on pen digits character recognition dataset

Three supervised learning methods on pen digits character recognition dataset Three supervsed learnng methods on pen dgts character recognton dataset Chrs Flezach Department of Computer Scence and Engneerng Unversty of Calforna, San Dego San Dego, CA 92093 cflezac@cs.ucsd.edu Satoru

More information

A Bootstrap Approach to Robust Regression

A Bootstrap Approach to Robust Regression Internatonal Journal of Appled Scence and Technology Vol. No. 9; November A Bootstrap Approach to Robust Regresson Dr. Hamadu Dallah Department of Actuaral Scence and Insurance Unversty of Lagos Akoka,

More information

Available online at ScienceDirect. Procedia Environmental Sciences 26 (2015 )

Available online at   ScienceDirect. Procedia Environmental Sciences 26 (2015 ) Avalable onlne at www.scencedrect.com ScenceDrect Proceda Envronmental Scences 26 (2015 ) 109 114 Spatal Statstcs 2015: Emergng Patterns Calbratng a Geographcally Weghted Regresson Model wth Parameter-Specfc

More information

Semi-Supervised Biased Maximum Margin Analysis for Interactive Image Retrieval

Semi-Supervised Biased Maximum Margin Analysis for Interactive Image Retrieval IP-06850-00.R3 Sem-Suervsed Based Maxmum Margn Analyss for Interactve Image Retreval Lnng Zhang,, Student Member, IEEE, Lo Wang, Senor Member, IEEE and Wes Ln 3, Senor Member, IEEE School of Electrcal

More information

Mixed Linear System Estimation and Identification

Mixed Linear System Estimation and Identification 48th IEEE Conference on Decson and Control, Shangha, Chna, December 2009 Mxed Lnear System Estmaton and Identfcaton A. Zymns S. Boyd D. Gornevsky Abstract We consder a mxed lnear system model, wth both

More information

Helsinki University Of Technology, Systems Analysis Laboratory Mat Independent research projects in applied mathematics (3 cr)

Helsinki University Of Technology, Systems Analysis Laboratory Mat Independent research projects in applied mathematics (3 cr) Helsnk Unversty Of Technology, Systems Analyss Laboratory Mat-2.08 Independent research projects n appled mathematcs (3 cr) "! #$&% Antt Laukkanen 506 R ajlaukka@cc.hut.f 2 Introducton...3 2 Multattrbute

More information

Unsupervised Learning and Clustering

Unsupervised Learning and Clustering Unsupervsed Learnng and Clusterng Why consder unlabeled samples?. Collectng and labelng large set of samples s costly Gettng recorded speech s free, labelng s tme consumng 2. Classfer could be desgned

More information

Modeling Inter-cluster and Intra-cluster Discrimination Among Triphones

Modeling Inter-cluster and Intra-cluster Discrimination Among Triphones Modelng Inter-cluster and Intra-cluster Dscrmnaton Among Trphones Tom Ko, Bran Mak and Dongpeng Chen Department of Computer Scence and Engneerng The Hong Kong Unversty of Scence and Technology Clear Water

More information

Determining the Optimal Bandwidth Based on Multi-criterion Fusion

Determining the Optimal Bandwidth Based on Multi-criterion Fusion Proceedngs of 01 4th Internatonal Conference on Machne Learnng and Computng IPCSIT vol. 5 (01) (01) IACSIT Press, Sngapore Determnng the Optmal Bandwdth Based on Mult-crteron Fuson Ha-L Lang 1+, Xan-Mn

More information

Support Vector Machines

Support Vector Machines Support Vector Machnes Decson surface s a hyperplane (lne n 2D) n feature space (smlar to the Perceptron) Arguably, the most mportant recent dscovery n machne learnng In a nutshell: map the data to a predetermned

More information

A Robust Method for Estimating the Fundamental Matrix

A Robust Method for Estimating the Fundamental Matrix Proc. VIIth Dgtal Image Computng: Technques and Applcatons, Sun C., Talbot H., Ourseln S. and Adraansen T. (Eds.), 0- Dec. 003, Sydney A Robust Method for Estmatng the Fundamental Matrx C.L. Feng and Y.S.

More information

Multilevel Analysis with Informative Weights

Multilevel Analysis with Informative Weights Secton on Survey Research Methods JSM 2008 Multlevel Analyss wth Informatve Weghts Frank Jenkns Westat, 650 Research Blvd., Rockvlle, MD 20850 Abstract Multlevel modelng has become common n large scale

More information

Announcements. Supervised Learning

Announcements. Supervised Learning Announcements See Chapter 5 of Duda, Hart, and Stork. Tutoral by Burge lnked to on web page. Supervsed Learnng Classfcaton wth labeled eamples. Images vectors n hgh-d space. Supervsed Learnng Labeled eamples

More information

C2 Training: June 8 9, Combining effect sizes across studies. Create a set of independent effect sizes. Introduction to meta-analysis

C2 Training: June 8 9, Combining effect sizes across studies. Create a set of independent effect sizes. Introduction to meta-analysis C2 Tranng: June 8 9, 2010 Introducton to meta-analyss The Campbell Collaboraton www.campbellcollaboraton.org Combnng effect szes across studes Compute effect szes wthn each study Create a set of ndependent

More information

Proper Choice of Data Used for the Estimation of Datum Transformation Parameters

Proper Choice of Data Used for the Estimation of Datum Transformation Parameters Proper Choce of Data Used for the Estmaton of Datum Transformaton Parameters Hakan S. KUTOGLU, Turkey Key words: Coordnate systems; transformaton; estmaton, relablty. SUMMARY Advances n technologes and

More information

Fuzzy Filtering Algorithms for Image Processing: Performance Evaluation of Various Approaches

Fuzzy Filtering Algorithms for Image Processing: Performance Evaluation of Various Approaches Proceedngs of the Internatonal Conference on Cognton and Recognton Fuzzy Flterng Algorthms for Image Processng: Performance Evaluaton of Varous Approaches Rajoo Pandey and Umesh Ghanekar Department of

More information

SIGGRAPH Interactive Image Cutout. Interactive Graph Cut. Interactive Graph Cut. Interactive Graph Cut. Hard Constraints. Lazy Snapping.

SIGGRAPH Interactive Image Cutout. Interactive Graph Cut. Interactive Graph Cut. Interactive Graph Cut. Hard Constraints. Lazy Snapping. SIGGRAPH 004 Interactve Image Cutout Lazy Snappng Yn L Jan Sun Ch-Keung Tang Heung-Yeung Shum Mcrosoft Research Asa Hong Kong Unversty Separate an object from ts background Compose the object on another

More information

For instance, ; the five basic number-sets are increasingly more n A B & B A A = B (1)

For instance, ; the five basic number-sets are increasingly more n A B & B A A = B (1) Secton 1.2 Subsets and the Boolean operatons on sets If every element of the set A s an element of the set B, we say that A s a subset of B, or that A s contaned n B, or that B contans A, and we wrte A

More information

A Robust LS-SVM Regression

A Robust LS-SVM Regression PROCEEDIGS OF WORLD ACADEMY OF SCIECE, EGIEERIG AD ECHOLOGY VOLUME 7 AUGUS 5 ISS 37- A Robust LS-SVM Regresson József Valyon, and Gábor Horváth Abstract In comparson to the orgnal SVM, whch nvolves a quadratc

More information

Environmental Modelling & Software

Environmental Modelling & Software Envronmental Modellng & Software 90 (2017) 68e77 Contents lsts avalable at ScenceDrect Envronmental Modellng & Software ournal homeage: www.elsever.com/locate/envsoft Addng satal flexblty to source-recetor

More information

Fuzzy Logic Based RS Image Classification Using Maximum Likelihood and Mahalanobis Distance Classifiers

Fuzzy Logic Based RS Image Classification Using Maximum Likelihood and Mahalanobis Distance Classifiers Research Artcle Internatonal Journal of Current Engneerng and Technology ISSN 77-46 3 INPRESSCO. All Rghts Reserved. Avalable at http://npressco.com/category/jcet Fuzzy Logc Based RS Image Usng Maxmum

More information

A Statistical Model Selection Strategy Applied to Neural Networks

A Statistical Model Selection Strategy Applied to Neural Networks A Statstcal Model Selecton Strategy Appled to Neural Networks Joaquín Pzarro Elsa Guerrero Pedro L. Galndo joaqun.pzarro@uca.es elsa.guerrero@uca.es pedro.galndo@uca.es Dpto Lenguajes y Sstemas Informátcos

More information

Parallelism for Nested Loops with Non-uniform and Flow Dependences

Parallelism for Nested Loops with Non-uniform and Flow Dependences Parallelsm for Nested Loops wth Non-unform and Flow Dependences Sam-Jn Jeong Dept. of Informaton & Communcaton Engneerng, Cheonan Unversty, 5, Anseo-dong, Cheonan, Chungnam, 330-80, Korea. seong@cheonan.ac.kr

More information

Application of Genetic Algorithms in Graph Theory and Optimization. Qiaoyan Yang, Qinghong Zeng

Application of Genetic Algorithms in Graph Theory and Optimization. Qiaoyan Yang, Qinghong Zeng 3rd Internatonal Conference on Materals Engneerng, Manufacturng Technology and Control (ICMEMTC 206) Alcaton of Genetc Algorthms n Grah Theory and Otmzaton Qaoyan Yang, Qnghong Zeng College of Mathematcs,

More information

Analysis of Malaysian Wind Direction Data Using ORIANA

Analysis of Malaysian Wind Direction Data Using ORIANA Modern Appled Scence March, 29 Analyss of Malaysan Wnd Drecton Data Usng ORIANA St Fatmah Hassan (Correspondng author) Centre for Foundaton Studes n Scence Unversty of Malaya, 63 Kuala Lumpur, Malaysa

More information

Skew Estimation in Document Images Based on an Energy Minimization Framework

Skew Estimation in Document Images Based on an Energy Minimization Framework Skew Estmaton n Document Images Based on an Energy Mnmzaton Framework Youbao Tang 1, Xangqan u 1, e Bu 2, and Hongyang ang 3 1 School of Comuter Scence and Technology, Harbn Insttute of Technology, Harbn,

More information

Improvement of Spatial Resolution Using BlockMatching Based Motion Estimation and Frame. Integration

Improvement of Spatial Resolution Using BlockMatching Based Motion Estimation and Frame. Integration Improvement of Spatal Resoluton Usng BlockMatchng Based Moton Estmaton and Frame Integraton Danya Suga and Takayuk Hamamoto Graduate School of Engneerng, Tokyo Unversty of Scence, 6-3-1, Nuku, Katsuska-ku,

More information

Extension of the Application of Conway-Maxwell-Poisson Models: Analyzing Traffic Crash Data Exhibiting Under-Dispersion

Extension of the Application of Conway-Maxwell-Poisson Models: Analyzing Traffic Crash Data Exhibiting Under-Dispersion Extenson of the Applcaton of Conway-Maxwell-Posson Models: Analyzng Traffc Crash Data Exhbtng Under-Dsperson Domnque Lord 1 Assstant Professor Zachry Department of Cvl Engneerng Texas A&M Unversty 3136

More information

NUMERICAL SOLVING OPTIMAL CONTROL PROBLEMS BY THE METHOD OF VARIATIONS

NUMERICAL SOLVING OPTIMAL CONTROL PROBLEMS BY THE METHOD OF VARIATIONS ARPN Journal of Engneerng and Appled Scences 006-017 Asan Research Publshng Network (ARPN). All rghts reserved. NUMERICAL SOLVING OPTIMAL CONTROL PROBLEMS BY THE METHOD OF VARIATIONS Igor Grgoryev, Svetlana

More information

Adjustment methods for differential measurement errors in multimode surveys

Adjustment methods for differential measurement errors in multimode surveys Adjustment methods for dfferental measurement errors n multmode surveys Salah Merad UK Offce for Natonal Statstcs ESSnet MM DCSS, Fnal Meetng Wesbaden, Germany, 4-5 September 2014 Outlne Introducton Stablsng

More information

Categories and Subject Descriptors B.7.2 [Integrated Circuits]: Design Aids Verification. General Terms Algorithms

Categories and Subject Descriptors B.7.2 [Integrated Circuits]: Design Aids Verification. General Terms Algorithms 3. Fndng Determnstc Soluton from Underdetermned Equaton: Large-Scale Performance Modelng by Least Angle Regresson Xn L ECE Department, Carnege Mellon Unversty Forbs Avenue, Pttsburgh, PA 3 xnl@ece.cmu.edu

More information

IMRT workflow. Optimization and Inverse planning. Intensity distribution IMRT IMRT. Dose optimization for IMRT. Bram van Asselen

IMRT workflow. Optimization and Inverse planning. Intensity distribution IMRT IMRT. Dose optimization for IMRT. Bram van Asselen IMRT workflow Otmzaton and Inverse lannng 69 Gy Bram van Asselen IMRT Intensty dstrbuton Webb 003: IMRT s the delvery of radaton to the atent va felds that have non-unform radaton fluence Purose: Fnd a

More information

MULTISPECTRAL IMAGES CLASSIFICATION BASED ON KLT AND ATR AUTOMATIC TARGET RECOGNITION

MULTISPECTRAL IMAGES CLASSIFICATION BASED ON KLT AND ATR AUTOMATIC TARGET RECOGNITION MULTISPECTRAL IMAGES CLASSIFICATION BASED ON KLT AND ATR AUTOMATIC TARGET RECOGNITION Paulo Quntlano 1 & Antono Santa-Rosa 1 Federal Polce Department, Brasla, Brazl. E-mals: quntlano.pqs@dpf.gov.br and

More information

Contour Error of the 3-DoF Hydraulic Translational Parallel Manipulator. Ryszard Dindorf 1,a, Piotr Wos 2,b

Contour Error of the 3-DoF Hydraulic Translational Parallel Manipulator. Ryszard Dindorf 1,a, Piotr Wos 2,b Advanced Materals Research Vol. 874 (2014) 57-62 Onlne avalable snce 2014/Jan/08 at www.scentfc.net (2014) rans ech Publcatons, Swtzerland do:10.4028/www.scentfc.net/amr.874.57 Contour Error of the 3-DoF

More information

MODERN animal and plant breeding schemes select

MODERN animal and plant breeding schemes select GENOMIC SELECTION WholeGenome Regresson and Predcton Methods Aled to Plant and Anmal Breedng Gustavo de los Camos* John M. Hckey Rcardo PongWong Hans D. Daetwyler and Maro P. L. Calus** *Deartment of Bostatstcs

More information

SVM-based Learning for Multiple Model Estimation

SVM-based Learning for Multiple Model Estimation SVM-based Learnng for Multple Model Estmaton Vladmr Cherkassky and Yunqan Ma Department of Electrcal and Computer Engneerng Unversty of Mnnesota Mnneapols, MN 55455 {cherkass,myq}@ece.umn.edu Abstract:

More information

Hybridization of Expectation-Maximization and K-Means Algorithms for Better Clustering Performance

Hybridization of Expectation-Maximization and K-Means Algorithms for Better Clustering Performance BULGARIAN ACADEMY OF SCIENCES CYBERNETICS AND INFORMATION TECHNOLOGIES Volume 16, No 2 Sofa 2016 Prnt ISSN: 1311-9702; Onlne ISSN: 1314-4081 DOI: 10.1515/cat-2016-0017 Hybrdzaton of Expectaton-Maxmzaton

More information

IX Price and Volume Measures: Specific QNA-ANA Issues

IX Price and Volume Measures: Specific QNA-ANA Issues IX Prce and Volume Measures: Secfc QNA-ANA Issues A. Introducton 9.. Ths chater addresses a selected set of ssues for constructng tme seres of rce and volume measures that are of secfc mortance for the

More information

Outline. Type of Machine Learning. Examples of Application. Unsupervised Learning

Outline. Type of Machine Learning. Examples of Application. Unsupervised Learning Outlne Artfcal Intellgence and ts applcatons Lecture 8 Unsupervsed Learnng Professor Danel Yeung danyeung@eee.org Dr. Patrck Chan patrckchan@eee.org South Chna Unversty of Technology, Chna Introducton

More information

Direct Analysis of Pre-Adjusted Loss Cost, Frequency or Severity in Tweedie Models

Direct Analysis of Pre-Adjusted Loss Cost, Frequency or Severity in Tweedie Models Drect Analyss of Pre-Adsted Loss Cost, Freqency or Severty n Tweede Models Sheng G. Sh, Ph.D. Abstract Resonse data (loss cost, clam freqency or clam severty are often re-adsted wth known factors and drectly

More information

Bayesian Networks: Independencies and Inference. What Independencies does a Bayes Net Model?

Bayesian Networks: Independencies and Inference. What Independencies does a Bayes Net Model? Bayesan Networks: Indeendences and Inference Scott Daves and Andrew Moore Note to other teachers and users of these sldes. Andrew and Scott would be delghted f you found ths source materal useful n gvng

More information

Statistical analysis on mean rainfall and mean temperature via functional data analysis technique

Statistical analysis on mean rainfall and mean temperature via functional data analysis technique Statstcal analyss on mean ranfall and mean temperature va functonal data analyss technque Jamaludn Suhala Ctaton: AIP Conference Proceedngs 1613, 368 (01); do: 10.1063/1.89361 Vew onlne: http://dx.do.org/10.1063/1.89361

More information

Recognizing Faces. Outline

Recognizing Faces. Outline Recognzng Faces Drk Colbry Outlne Introducton and Motvaton Defnng a feature vector Prncpal Component Analyss Lnear Dscrmnate Analyss !"" #$""% http://www.nfotech.oulu.f/annual/2004 + &'()*) '+)* 2 ! &

More information

An Entropy-Based Approach to Integrated Information Needs Assessment

An Entropy-Based Approach to Integrated Information Needs Assessment Dstrbuton Statement A: Approved for publc release; dstrbuton s unlmted. An Entropy-Based Approach to ntegrated nformaton Needs Assessment June 8, 2004 Wllam J. Farrell Lockheed Martn Advanced Technology

More information

Cluster Analysis of Electrical Behavior

Cluster Analysis of Electrical Behavior Journal of Computer and Communcatons, 205, 3, 88-93 Publshed Onlne May 205 n ScRes. http://www.scrp.org/ournal/cc http://dx.do.org/0.4236/cc.205.350 Cluster Analyss of Electrcal Behavor Ln Lu Ln Lu, School

More information

User Authentication Based On Behavioral Mouse Dynamics Biometrics

User Authentication Based On Behavioral Mouse Dynamics Biometrics User Authentcaton Based On Behavoral Mouse Dynamcs Bometrcs Chee-Hyung Yoon Danel Donghyun Km Department of Computer Scence Department of Computer Scence Stanford Unversty Stanford Unversty Stanford, CA

More information

A Similarity-Based Prognostics Approach for Remaining Useful Life Estimation of Engineered Systems

A Similarity-Based Prognostics Approach for Remaining Useful Life Estimation of Engineered Systems 2008 INTERNATIONAL CONFERENCE ON PROGNOSTICS AND HEALTH MANAGEMENT A Smlarty-Based Prognostcs Approach for Remanng Useful Lfe Estmaton of Engneered Systems Tany Wang, Janbo Yu, Davd Segel, and Jay Lee

More information

Optimized Query Planning of Continuous Aggregation Queries in Dynamic Data Dissemination Networks

Optimized Query Planning of Continuous Aggregation Queries in Dynamic Data Dissemination Networks WWW 007 / Trac: Performance and Scalablty Sesson: Scalable Systems for Dynamc Content Otmzed Query Plannng of Contnuous Aggregaton Queres n Dynamc Data Dssemnaton Networs Rajeev Guta IBM Inda Research

More information

Parameter estimation for incomplete bivariate longitudinal data in clinical trials

Parameter estimation for incomplete bivariate longitudinal data in clinical trials Parameter estmaton for ncomplete bvarate longtudnal data n clncal trals Naum M. Khutoryansky Novo Nordsk Pharmaceutcals, Inc., Prnceton, NJ ABSTRACT Bvarate models are useful when analyzng longtudnal data

More information

6.854 Advanced Algorithms Petar Maymounkov Problem Set 11 (November 23, 2005) With: Benjamin Rossman, Oren Weimann, and Pouya Kheradpour

6.854 Advanced Algorithms Petar Maymounkov Problem Set 11 (November 23, 2005) With: Benjamin Rossman, Oren Weimann, and Pouya Kheradpour 6.854 Advanced Algorthms Petar Maymounkov Problem Set 11 (November 23, 2005) Wth: Benjamn Rossman, Oren Wemann, and Pouya Kheradpour Problem 1. We reduce vertex cover to MAX-SAT wth weghts, such that the

More information

Control strategies for network efficiency and resilience with route choice

Control strategies for network efficiency and resilience with route choice Control strateges for networ effcency and reslence wth route choce Andy Chow Ru Sha Centre for Transport Studes Unversty College London, UK Centralsed strateges UK 1 Centralsed strateges Some effectve

More information

TIME-EFFICIENT NURBS CURVE EVALUATION ALGORITHMS

TIME-EFFICIENT NURBS CURVE EVALUATION ALGORITHMS TIME-EFFICIENT NURBS CURVE EVALUATION ALGORITHMS Kestuts Jankauskas Kaunas Unversty of Technology, Deartment of Multmeda Engneerng, Studentu st. 5, LT-5368 Kaunas, Lthuana, kestuts.jankauskas@ktu.lt Abstract:

More information

Solving two-person zero-sum game by Matlab

Solving two-person zero-sum game by Matlab Appled Mechancs and Materals Onlne: 2011-02-02 ISSN: 1662-7482, Vols. 50-51, pp 262-265 do:10.4028/www.scentfc.net/amm.50-51.262 2011 Trans Tech Publcatons, Swtzerland Solvng two-person zero-sum game by

More information

Air Transport Demand. Ta-Hui Yang Associate Professor Department of Logistics Management National Kaohsiung First Univ. of Sci. & Tech.

Air Transport Demand. Ta-Hui Yang Associate Professor Department of Logistics Management National Kaohsiung First Univ. of Sci. & Tech. Ar Transport Demand Ta-Hu Yang Assocate Professor Department of Logstcs Management Natonal Kaohsung Frst Unv. of Sc. & Tech. 1 Ar Transport Demand Demand for ar transport between two ctes or two regons

More information

Active Contours/Snakes

Active Contours/Snakes Actve Contours/Snakes Erkut Erdem Acknowledgement: The sldes are adapted from the sldes prepared by K. Grauman of Unversty of Texas at Austn Fttng: Edges vs. boundares Edges useful sgnal to ndcate occludng

More information

Machine Learning. Support Vector Machines. (contains material adapted from talks by Constantin F. Aliferis & Ioannis Tsamardinos, and Martin Law)

Machine Learning. Support Vector Machines. (contains material adapted from talks by Constantin F. Aliferis & Ioannis Tsamardinos, and Martin Law) Machne Learnng Support Vector Machnes (contans materal adapted from talks by Constantn F. Alfers & Ioanns Tsamardnos, and Martn Law) Bryan Pardo, Machne Learnng: EECS 349 Fall 2014 Support Vector Machnes

More information

A new Algorithm for Lossless Compression applied to two-dimensional Static Images

A new Algorithm for Lossless Compression applied to two-dimensional Static Images A new Algorthm for Lossless Comresson aled to two-dmensonal Statc Images JUAN IGNACIO LARRAURI Deartment of Technology Industral Unversty of Deusto Avda. Unversdades, 4. 48007 Blbao SPAIN larrau@deusto.es

More information

Application of Neural Network Classifiers to Disease Diagnosis

Application of Neural Network Classifiers to Disease Diagnosis Alcaton of Neural Network Classfers to Dsease Dagnoss Mural Shanker Deartment of ADMS, College of Busness Kent State Unversty, Kent OH 44242-0001 (216) 672-2750 Ext. 347 mshanker@scoro.kent.edu Abstract

More information

Outlier Detection based on Robust Parameter Estimates

Outlier Detection based on Robust Parameter Estimates Outler Detecton based on Robust Parameter Estmates Nor Azlda Aleng 1, Ny Ny Nang, Norzan Mohamed 3 and Kasyp Mokhtar 4 1,3 School of Informatcs and Appled Mathematcs, Unverst Malaysa Terengganu, 1030 Kuala

More information

GSLM Operations Research II Fall 13/14

GSLM Operations Research II Fall 13/14 GSLM 58 Operatons Research II Fall /4 6. Separable Programmng Consder a general NLP mn f(x) s.t. g j (x) b j j =. m. Defnton 6.. The NLP s a separable program f ts objectve functon and all constrants are

More information

SOFT COMPUTING OPTIMIZER FOR INTELLIGENT CONTROL SYSTEMS DESIGN: THE STRUCTURE AND APPLICATIONS

SOFT COMPUTING OPTIMIZER FOR INTELLIGENT CONTROL SYSTEMS DESIGN: THE STRUCTURE AND APPLICATIONS SOFT COMPUTING OPTIMIZER FOR INTELLIGENT CONTROL SYSTEMS DESIGN: THE STRUCTURE AND APPLICATIONS Sergey A. PANFILOV, Ludmla V. LITVINTSEVA, Ilya S. ULYANOV, Kazuk TAKAHASHI, Sergue V. ULYANOV YAMAHA Motor

More information

CS246: Mining Massive Datasets Jure Leskovec, Stanford University

CS246: Mining Massive Datasets Jure Leskovec, Stanford University CS46: Mnng Massve Datasets Jure Leskovec, Stanford Unversty http://cs46.stanford.edu /19/013 Jure Leskovec, Stanford CS46: Mnng Massve Datasets, http://cs46.stanford.edu Perceptron: y = sgn( x Ho to fnd

More information

High resolution 3D Tau-p transform by matching pursuit Weiping Cao* and Warren S. Ross, Shearwater GeoServices

High resolution 3D Tau-p transform by matching pursuit Weiping Cao* and Warren S. Ross, Shearwater GeoServices Hgh resoluton 3D Tau-p transform by matchng pursut Wepng Cao* and Warren S. Ross, Shearwater GeoServces Summary The 3D Tau-p transform s of vtal sgnfcance for processng sesmc data acqured wth modern wde

More information

Broadcast Time Synchronization Algorithm for Wireless Sensor Networks Chaonong Xu 1)2)3), Lei Zhao 1)2), Yongjun Xu 1)2) and Xiaowei Li 1)2)

Broadcast Time Synchronization Algorithm for Wireless Sensor Networks Chaonong Xu 1)2)3), Lei Zhao 1)2), Yongjun Xu 1)2) and Xiaowei Li 1)2) Broadcast Tme Synchronzaton Algorthm for Wreless Sensor Networs Chaonong Xu )2)3), Le Zhao )2), Yongun Xu )2) and Xaowe L )2) ) Key Laboratory of Comuter Archtecture, Insttute of Comutng Technology Chnese

More information

Classification / Regression Support Vector Machines

Classification / Regression Support Vector Machines Classfcaton / Regresson Support Vector Machnes Jeff Howbert Introducton to Machne Learnng Wnter 04 Topcs SVM classfers for lnearly separable classes SVM classfers for non-lnearly separable classes SVM

More information

Machine Learning: Algorithms and Applications

Machine Learning: Algorithms and Applications 14/05/1 Machne Learnng: Algorthms and Applcatons Florano Zn Free Unversty of Bozen-Bolzano Faculty of Computer Scence Academc Year 011-01 Lecture 10: 14 May 01 Unsupervsed Learnng cont Sldes courtesy of

More information

Unsupervised Learning

Unsupervised Learning Pattern Recognton Lecture 8 Outlne Introducton Unsupervsed Learnng Parametrc VS Non-Parametrc Approach Mxture of Denstes Maxmum-Lkelhood Estmates Clusterng Prof. Danel Yeung School of Computer Scence and

More information

Solitary and Traveling Wave Solutions to a Model. of Long Range Diffusion Involving Flux with. Stability Analysis

Solitary and Traveling Wave Solutions to a Model. of Long Range Diffusion Involving Flux with. Stability Analysis Internatonal Mathematcal Forum, Vol. 6,, no. 7, 8 Soltary and Travelng Wave Solutons to a Model of Long Range ffuson Involvng Flux wth Stablty Analyss Manar A. Al-Qudah Math epartment, Rabgh Faculty of

More information

Stitching of off-axis sub-aperture null measurements of an aspheric surface

Stitching of off-axis sub-aperture null measurements of an aspheric surface Sttchng of off-axs sub-aperture null measurements of an aspherc surface Chunyu Zhao* and James H. Burge College of optcal Scences The Unversty of Arzona 1630 E. Unversty Blvd. Tucson, AZ 85721 ABSTRACT

More information

5 The Primal-Dual Method

5 The Primal-Dual Method 5 The Prmal-Dual Method Orgnally desgned as a method for solvng lnear programs, where t reduces weghted optmzaton problems to smpler combnatoral ones, the prmal-dual method (PDM) has receved much attenton

More information

SLAM Summer School 2006 Practical 2: SLAM using Monocular Vision

SLAM Summer School 2006 Practical 2: SLAM using Monocular Vision SLAM Summer School 2006 Practcal 2: SLAM usng Monocular Vson Javer Cvera, Unversty of Zaragoza Andrew J. Davson, Imperal College London J.M.M Montel, Unversty of Zaragoza. josemar@unzar.es, jcvera@unzar.es,

More information

Synthesizer 1.0. User s Guide. A Varying Coefficient Meta. nalytic Tool. Z. Krizan Employing Microsoft Excel 2007

Synthesizer 1.0. User s Guide. A Varying Coefficient Meta. nalytic Tool. Z. Krizan Employing Microsoft Excel 2007 Syntheszer 1.0 A Varyng Coeffcent Meta Meta-Analytc nalytc Tool Employng Mcrosoft Excel 007.38.17.5 User s Gude Z. Krzan 009 Table of Contents 1. Introducton and Acknowledgments 3. Operatonal Functons

More information

3D FCRM MODELING IN MILES PER GALLON OF CAR

3D FCRM MODELING IN MILES PER GALLON OF CAR 3D FCRM Modelng n Mles Per Gallon of Car 3 4 3D FCRM MODELING IN MILES PER GALLON OF CAR Mohd Safullah Rusman Robah Adnan Efend N. Nasbo The new fuzzy c-regresson modelng (FcRM) are wdely used n order

More information

Image Segmentation. Image Segmentation

Image Segmentation. Image Segmentation Image Segmentaton REGION ORIENTED SEGMENTATION Let R reresent the entre mage regon. Segmentaton may be vewed as a rocess that arttons R nto n subregons, R, R,, Rn,such that n= R = R.e., the every xel must

More information