Statistics and Data Analysis. Use of the ROC Curve and the Bootstrap in Comparing Weighted Logistic Regression Models

Size: px
Start display at page:

Download "Statistics and Data Analysis. Use of the ROC Curve and the Bootstrap in Comparing Weighted Logistic Regression Models"

Transcription

1 Paper Use of the ROC Curve and the Bootstrap n Comparng Weghted Logstc Regresson Models Davd Izrael, Annabella A. Battagla, Davd C. Hoagln, and Mchael P. Battagla, Abt Assocates Inc., Cambrdge, MA Abstract In analyzng data from a survey, researchers often need to compare the effectveness of several logstc regresson models. The recever operatng characterstc curve offers one way to measure effectveness of predcton, by calculatng the area under the curve (AUC). We present a SAS macro for calculatng AUC that takes the survey weghts nto account. For comparng logstc regresson models, one needs to assess dfferences n AUC aganst the varaton n the data. We demonstrate the use of the SAS SURVEYSELECT procedure to create a set of 1,000 bootstrap samples and gve some background on the calculaton of separate weghts for each bootstrap sample. For each sample, the AUC macro s then used to calculate the AUC for each model. We show how to use the bootstrap results to assess the sgnfcance of the dfference n predctve ablty of the two models. 1. Introducton In analyzng data from a survey, we needed to compare the effectveness of several logstc regresson models. The recever operatng characterstc (ROC) curve offers one way to measure effectveness of predcton, by calculatng AUC. Then, for the comparsons, we needed to assess the dfferences n AUC aganst the varaton n the data. We had already developed a substantal set of bootstrap samples, and those allowed us to calculate a bootstrap standard error for the dfference n AUC, wthout makng any dstrbutonal assumptons. Wth ths bref overvew we now descrbe these components and then dscuss the applcaton of them n our study. A key ngredent s the samplng weghts assocated wth the survey data. The recever operatng characterstc curve s often used to descrbe the accuracy of tests n dagnostc medcne, as summarzed n the revew by Pepe (2000). Brefly, the test yelds a numercal result X, such that larger values are more ndcatve of dsease. One can choose a threshold z and dchotomze the test by defnng X z as a postve result. From subjects whose true dsease status s known (both dseased and nondseased), one obtans the falsepostve rate and the false-negatve rate for each value of z. The ROC curve s obtaned by plottng 1 mnus the falsenegatve rate aganst the false-postve rate for all possble choces of z. That s, each value of z yelds a pont on the curve, whch ncludes the pont (0,0) (f z s hgh enough, the test produces no postves) and the pont (1,1) (f z s low enough, all outcomes are postve). The area under the ROC curve provdes a summary of the accuracy of the dagnostc test. As Pepe ponts out, the AUC can be nterpreted as the probablty that the test result from a randomly chosen dseased ndvdual s more ndcatve of dsease than that from a randomly chosen nondseased ndvdual. Ths nterpretaton or equvalence, dscussed also by Hanley and McNel (1982), focuses attenton on the dstrbutons of the test result (for example, the concentraton of a chemcal n blood) n dseased and nondseased persons. If the two dstrbutons are clearly separated, the probablty wll be close to 1; but f they are centered at the same value, the probablty wll be ½. In the context of logstc regresson we refer to event cases and non-event cases, rather than dseased and nondseased persons. The test result s the predcted probablty of an event, from the logstc regresson model. The bootstrap (Efron 1982) uses resamplng to provde a bass for studyng the behavor of estmates. For a smple random sample of sze n, wth observatons x1, x2,..., x n, the man steps nvolve settng B (the number of bootstrap samples, usually large); usng samplng wth replacement * * * to draw a bootstrap sample of n, X1, X2,..., Xn, from the set { x 1, x 2,..., x n } (B tmes, ndependently); and calculatng the estmate, t, from each bootstrap sample to obtan * * * * t1, t2,..., t B. Analyss of the tb then yelds nformaton on the samplng dstrbuton of t when the data come from the populaton that underles x 1,..., x n. For example, the sample standard devaton of the t * b s the bootstrap standard error of t. When the data are a szable sample from a survey wth survey weghts, both the calculaton of the area under the ROC curve and applcaton of the bootstrap requre consderable specal programmng. Secton 2 dscusses the use of SAS to calculate AUC n the presence of survey weghts. Secton 3 comments on comparng the predctve value of logstc regresson models. Secton 4 sketches the basc framework for applyng the bootstrap to a complex sample survey, and Secton 5 llustrates the use of PROC SURVEYSELECT to create bootstrap samples. When the survey weghts nvolve adjustments, the elements of a bootstrap sample cannot smply nhert the weghts that they had n the orgnal sample. Secton 6 dscusses the 1

2 need to recalculate weghts, so that each bootstrap sample has ts own complete set of replcate weghts. Secton 7 then reports on the use of the bootstrap to estmate the standard error of AUC for a logstc regresson model and the standard error of the dfference n AUC between two such models. Fnally, Secton 8 adds some concludng dscusson. 2. Usng SAS to Estmate the Area under the ROC Curve We often use PROC LOGISTIC to ft logstc regresson models to weghted survey data. In fttng a model, PROC LOGISTIC takes the survey weghts nto account, but t gnores them n calculatng the ngredents of the ROC curve. Those ngredents are stored n the OUTROC data set, whch keeps one record for each dstnct predcted probablty and has the followng varables (whose values correspond to usng that probablty as the threshold): _POS_ - the number of correctly predcted event responses; _ NEG_ - the number of correctly predcted non-event responses; _FALPOS_ - the number of falsely predcted event responses; _FALNEG_ - the number of falsely predcted non-event responses; _SENSIT_ - the senstvty, whch s the proporton of event observatons that were predcted to have an event response; and _1MSPEC_ - 1 mnus specfcty, whch s the proporton of non-event observatons that were predcted to have an event response. In the presence of survey weghts, the varables for the ROC curve are not computed correctly and look exactly the same as f there were no weghts. The predcted probabltes, however, are correct. To calculate the AUC n the presence of survey weghts, we wrote a macro, CALCAUC, whch takes the weghts nto account when calculatng the varables for the ROC curve. We gve an overvew of the macro below and consder ts applcaton, both as a stand-alone program and as a subroutne n a bootstrap procedure. Macro CALCAUC Algorthm The macro calculates the varables for the ROC curve and then the AUC n the presence of survey weghts. To formalze the algorthm, we ncorporate weghts n the defntons of the ROC curve varables descrbed n Chapter 39, The Logstc Procedure, of SAS/STAT documentaton (Verson 8). Let the weghted number of ndvduals n a sample havng a certan event be n 1. Let ths group be denoted by C 1, and let the group of the remanng n 2 (weghted) ndvduals who do not have the event be denoted by C 2. Let ˆp be an estmated probablty of the event n the weghted model. W( ) denotes the weghted ndcator functon. For example, f pˆ z, W( pˆ z) s the samplng weght of ndvdual. For each cutpont z, _POS_ (z) = Σ W C 1 _FALPOS_ (z) = Σ W ( p ˆ z ) C 2 (1) ( p ˆ z ) (2) _SENSIT_(z) = _POS_(z)/n 1 (3) 1MSPEC_(z) = _FALPOS_(z)/ n 2 (4) Note that _POS_ (z) s the weghted number of correctly predcted event responses, _FALPOS_ (z) s the weghted number of falsely predcted event responses, _SENSIT_(z) s the weghted senstvty of the model, and _1MSPEC_(z) s one mnus the weghted specfcty of the model. Havng calculated _SENSIT_ and _1MSPEC_, we use them to calculate the AUC as the sum of the area of trapezods. Formally, f S s a set of cutponts joned wth 0 as the ntal one and 1 as the last one, the AUC can be expressed by the formula : AUC =.5 ( _1MSPEC_ +1 - _1MSPEC_ )(_SENSIT_ +1 +_SENSIT_ ) S Exhbt 1 presents the macro (wth lne numbers). We now descrbe ts functons secton by secton and dscuss such ssues as computatonal effcency and resource consumpton. Overvew of the code Lnes 1 23 contan the macro s nput parameters; model represents the strng of explanatory varables, all of whch must be categorcal (otherwse the number of dstnct predcted probabltes could be very large); depvar s a response varable, assumed to have the value of 1 for an event and 0 for a non-event; round and acceler control effcency of the macro and wll be descrbed below; replca must be blank when runnng the macro as a stand-alone program; otherwse t must be assgned the name of a macro varable that serves as a replcate counter when calculaton of AUC s done for each bootstrap replcate. Lnes check that the varables n the model are present n the nput data set. If not, the macro outputs (5) 2

3 names of absent varables nto the LOG and stops (Lnes 45 and 60). Lnes represent PROC LOGISTIC s statements and optons. The data set specfed n the OUTROC opton wll contan the dstnct estmated probabltes (_PROB_), whch wll serve as the cutponts mentoned n the Algorthm secton. Also, the data set _PROBS specfed n the opton OUT wll nclude all varables of the nput data set, along wth the predcted probablty _P_HAT. Lnes contan optonal statements that are ntended to accelerate the computatonal process. Computng tme s especally senstve to the number of dstnct estmated probabltes. We suggest reducng computng tme by roundng the predcted probabltes (lnes 76 and 85). The mpact of the roundng on the precson of the calculated AUC s ordnarly mnmal: we observed a dfference only n the ffth decmal place. Lnes restore the orgnal descendng order of the predcted probabltes, as roundng could change the orderng. Lnes calculate the weghted varables assocated wth the ROC curve, _POS_ and _FALPOS_ n partcular, followng formulas (1) and (2). The outer DO-loop (lne 92) sets sequentally the dstnct predcted probablty from OUTROC data set and passes t through the whole _PROBS data set, whch s accessed drectly n the nner DO-loop (lne 98). Lnes calculate _SENSIT_ and _1MSPEC_ accordng to formulas (3) and (4). Lnes calculate the AUC by formula (5). If we are computng the AUC for each bootstrap replcate, the name of the output data set wth the calculated value of AUC contans the replcate number. In ths stuaton the name of the varable wth the calculated area contans the replcate number as well (lne 141). Exhbt 1: CALCAUC Macro 1 %macro calcauc(dsanal =, /* INPUT DATA SET */ 2 3 outds = c, /* OUTPUT DATA SET WITH AUC*/ 4 5 d =, /* ID VARIABLE */ 6 7 weght =, /* SURVEY WEIGHT */ 8 9 model =, /* ALL EXPLANATORY VAR's. */ 10 /* MUST BE CATEGORICAL */ depvar =, /* DEPENDENT VARIABLE */ round =.001, /* PRECISION AT WHICH TO */ 15 /* ROUND PRED PROBABIL */ replca =, /* COUNTER OF REPLICATES */ 18 /* WHEN BOOTSTRAP IS USED */ acceler = y ); /* ACCELERATE CALCULATIONS 21 BY ROUNDING PREDICTED 22 PROBABILITIES. &ROUND 23 MUST BE PRESENT */ %let control = 1; % macro check; %local dsd control nullstr rc varnum; %let model=%upcase(&model); 32 %let depvar=%upcase(&depvar); 33 %let strng=&model &depvar; %let =1; 36 %let nullstr=; %let dsd=%sysfunc(open(&dsanal)); %do %untl(%scan(&strng,&)=&nullstr); 41 %let varnum=%sysfunc(varnum(&dsd,%scan(&strng,&))); 42 %f &varnum=0 %then %do; 43 %let control=0; 44 %put ; 45 %put VARIABLE %scan(&strng,&) APPEARS IN THE 46 MODEL, BUT NOT IN THE INPUT DATA SET; 47 %put ; 48 %end; 49 %let =%eval(&+1); 50 %end; %let rc=%sysfunc(close(&dsd)); 53 %mend check; %f (&replca=) or (&replca=1) %then %check; %f &control = 0 %then %do; 58 %put **** MACRO TERMINATED BECAUSE OF ERRORS 59 ABOVE ******; 60 %goto ext; 61 %end; 62 %else %do; proc logstc descendng data=&dsanal; 65 weght &weght /norm; 66 class &model; 67 model &depvar= &model/ 68 outroc=_roc(keep=_prob_); 69 output out=_probs predcted=_p_hat; %f %upcase(&acceler) = Y %then %do; data _roc; 75 set _roc; 76 _prob_=round(_prob_, &round); proc sort nodupkey; 80 by descendng _prob_; data _probs; 84 set _probs; 85 _p_hat=round(_p_hat,&round); 86 3

4 87 88 %end; data _out1 (keep= _pos neg falpos falneg_); do =1 to numobroc; set roc1 nobs=numobroc pont= ; 95 retan _pos neg falpos falneg_ ; 96 pos=0; neg=0; falpos=0; falneg=0; do j=1 to numobpro; set probs nobs=numobpro pont=j; 101 f _p_hat >=_prob_ then _preddep=1; else _preddep=0; 102 f &depvar=1 and _preddep=1 then _pos_=_pos_+ &weght; 103 else 104 f &depvar=0 and _preddep=1 then _falpos_=_falpos_+ &weght; 105 else 106 f &depvar=1 and _preddep=0 then _falneg_=_falneg_+&weght; 107 else 108 _neg_=_neg_+&weght; f j = numobpro then output; end; 113 end; 114 stop; data _s; 118 set _out1; 119 f _n_=1 then set _out1(rename=(_pos_=_n1 120 _falpos_=_n2)) nobs=numout pont=numout; _senst_=_pos_/_n1; 123 _1mspec_=_falpos_/_n2; data &outds&replca(keep=area&replca); 127 set _s end=fn; 128 retan _w _z area&replca 0; 129 f _n_=1 then do; _w=0; _z=0; 130 end; 131 _x=_1mspec_-_w; 132 _y=(_senst_+_z)*0.5; 133 _z=_senst_; 134 _w=_1mspec_; area&replca=sum(area&replca,_x*_y); f fn then output; proc prnt data=&outds&replca; %end; 144 %ext:; 145 %mend calcauc; 3. Comparng the Predctve Value of Two Models Many analyses nvolve fttng two or more logstc regresson models to the same data. Then, n choosng among the models, t may be useful to compare ther predctve value, va the AUC. Often one of the models s the fnal model or full model from a stepwse logstc regresson, and the other models are subsets of the full model (e.g., the frst k predctors to enter a reduced model). The dfference between the AUC for the full model and the AUC for a reduced model can ad n judgng whether the full model offers a real advantage over the reduced model. (Ths applcaton of the dfference n AUC does not requre that the models be nested. It s applcable to the comparson of any two models.) To assess the sze of the dfference n AUC relatve to the varaton n the data, we need the estmated standard error of the dfference. One sutable approach s the bootstrap method, whch uses replcaton (Wolter 1985). As mentoned n Secton 1, the bootstrap nvolves drawng repeated ndependent samples (wth replacement) from the orgnal sample and then estmatng the AUC for each model and the dfference n AUC, for each of these bootstrap samples. The sample standard devaton of that dfference (over the bootstrap samples) s the bootstrap estmate of ts standard error. 4. The Bootstrap Method for Varance Estmaton Our data came from a stratfed one-stage cluster sample of over 20,000 persons that ncorporates several weghtng adjustments. The sample desgn entals stratfcaton of the U.S. nto 78 geographc areas. Wthn each stratum, a random sample of households s drawn. The survey collects data on all elgble household members, makng households the clusters n the sample desgn. Rust and Rao (1996) dscuss the use of replcaton methods to obtan standard errors for complex survey desgns. The applcaton of the bootstrap procedure to our sample desgn nvolves drawng the bootstrap samples (replcates) wthn each stratum. In connecton wth other analyses of the same data, we had prevously constructed 1,000 bootstrap replcates, n order to obtan bounds for 95% confdence ntervals drectly from the dstrbuton of the bootstrap estmates, as well as bootstrap standard errors. Thus, t was natural to use those 1,000 bootstrap replcates n estmatng the standard error of the dfference n AUC. The next secton descrbes the use of PROC SURVEYSELECT to construct bootstrap replcates. Then Secton 6 dscusses the further steps requred to produce samplng weghts specfc to each replcate. 5. Use of PROC SURVEYSELECT The followng statements show how the SURVEYSELECT procedure was used to draw the 1,000 bootstrap replcates. %let dd = ourdrectory; %let n_ter=1; %let max_ter=1000; %let n_smpfle=&dd..samplefle; %let n_nsze= geo_area_tot; /* CREATE A DATASET WITH SAMPLE SIZES TO BE DRAWN FROM EACH STRATUM */ 4

5 proc freq data=&n_smpfle; table geo_area/out=&n_nsze(rename=(count=_nsze_) drop=percent); /* MACRO TO DRAW 1000 BOOTSTRAP SAMPLES */ %MACRO BOOTREP; %do =&n_ter %to &max_ter; proc prntto new prnt="brep_&..lst"; proc prntto new log="brep_&..log"; ttle3 " REPLICATE =& URS selecton"; optons pageno=1; proc surveyselect data=&n_smpfle method=urs sampsze=&n_nsze out=&dd..urs_& outhts; strata geo_area; proc freq data=&dd..urs_&; tables NumberHts; %end; %MEND BOOTREP; %BOOTREP To draw the 1,000 sample replcates wth equal probablty and wth replacement, we used METHOD=URS (Unrestrcted Random Samplng). SAMPSIZE = GEO_AREA_TOT dentfes the SAS data set that contans _NSIZE_, the dfferent sample szes for the strata. The OUT=&dd..urs_& opton outputs each of the 1,000 samples nto a separate permanent SAS dataset. The OUTHITS opton outputs a separate observaton for each selecton when an observaton s selected more than once. The output dataset contans for each observaton the varable NumberHts, the number of tmes a household was selected nto the sample n a gven replcate. The STRATA statement defnes the varable GEO_AREA as the stratfcaton varable. 6. Calculaton of Replcate Weghts Rust and Rao (1986) gve a method for adjustng the fnal samplng weghts to obtan bootstrap weghts. They also note, however, that for the varance estmators to reman close to unbased, the weght adjustment steps appled to the orgnal sample should be appled to each bootstrap replcate. Ths s an mportant consderaton n our sample desgn, gven the consderable number of weght adjustments. Thus, for each bootstrap replcate we repeated all of the weght calculaton steps. As a result each of the 1,000 bootstrap replcates has ts own set of weghts. 7. Usng SAS and the Bootstrap Replcates to Estmate the Varance of the AUC Applyng the macro CALCAUC to the orgnal sample, we calculate the AUC for the weghted models wth 14 explanatory varables (full model) and 6 explanatory varables (reduced model). We denote them by AUC14 and AUC6, respectvely. In Exhbt 2 the macro ALLREPL uses the macro CALCAUC as a subroutne to reft a weghted logstc regresson model and obtan the AUC for each of 1,000 bootstrap replcates. Exhbt 2: ALLREPL Macro %let youranal = anal; /*ANALYTIC FILE WITH ALL DATA */ %let dsbswts = replwts; /* DATA SET WITH ID AND 1,000 REPLICATE WEIGHTS */ %let model = yourmodel; /* STRING WITH EXPLANAT ORY VARIABLES */ %let depvar = yourresponse; /* RESPONSE VARIABLE */ %macro allrepl (start,end); %do v=&start %to &end ; /* &START and &END ARE FIRST AND LAST REPLICATE TO PROCESS, 1 AND 1,000 IN OUR EXAMPLE */ data _anal; merge &youranal (n=_1 ) &dsbswts (keep=id w&v where=(w&v ne 0) n=_2); /* RETRIEVE &V-th REPLICATE WHERE V-th REPLICATE WEIGHT NE ZERO */ by ID; f _2; wgt=w&v; drop w&v; %calcauc( dsanal = _ANAL, /* CALCULATE AUC FOR &V-th REPLICATE */ outds = C, d = ID, weght = WGT, model = &YOURMODEL, /* REFIT MODEL TO DATA IN &V-th REPLICATE*/ depvar = &YOURESPONSE, round =.001, replca = &V, acceler = Y ); %end; %mend; %allrepl(1,1000) Applyng the ALLREPL macro for the full model and then for the reduced model, we obtan two sets of replcate AUCs: AUCR14 ( = 1 to 1,000), and AUCR6 ( = 1 to 1,000), respectvely. We denote the bootstrap sample AUC by AUCR to dstngush t from the one of the orgnal sample. To estmate the bootstrap standard errors of AUC14 and AUC6, we smply apply PROC UNIVARIATE to the AUCR14 and the AUCR6 ( = 1 to 1,000) to obtan the standard devatons. To estmate the standard error of the dfference n AUC between the two models, DIFF = AUC14 - AUC6, we apply PROC 5

6 UNIVARIATE to the dfferences AUCR14 - AUCR6 ( = 1 to 1,000) to obtan the standard devaton STDDIFF. Then t = abs(diff / STDDIFF) The correspondng sgnfcance level of a two-taled t test s gven by p = (1-PROBT(t, df))*2, The weghted AUC s.658 for the full model and.641 for the reduced model. The dfference n weghted AUC of.017 s hghly sgnfcant. Table 1 gves the results. Table 1: AUCs and Bootstrap Standard Errors Number Area or Dfference n Area Under the Curve of Area or Std. Err. Varance t p Predctors Dfference vs Dscusson The area under the recever operatng characterstc curve s an mportant and wdely used measure of the predctve ablty of a logstc regresson model. Most survey data fles have survey weghts attached. The LOGISTIC procedure does not take the weghts nto account n ts calculaton of the area under the ROC curve and therefore usually does not gve the correct value. The SAS macro CALCAUC uses the survey weghts n the calculaton of the area under the ROC curve by summng the area of trapezods. Hanley and McNel (1982) ndcate that the c statstc (whch PROC LOGISTIC reports n the Assocaton of Predcted Probabltes and Observed Responses table) s equvalent to the area under the ROC curve. We have also developed a SAS macro, not dscussed n ths paper, that calculates a weghted verson of the c statstc. We have used the area under the curve to compare the predctve ablty of two logstc regresson models estmated from the same survey data fle. Usng bootstrap samples, t s possble to test whether the two models have the same area under the ROC curve. We provde some background on how SURVEYSELECT can be used to create bootstrap samples. It s possble, however, that the survey fle beng analyzed already contans bootstrap samples and bootstrap replcate weghts. If not, t s wse to consult wth a statstcan who s famlar wth the bootstrap method of varance estmaton before creatng bootstrap samples and bootstrap replcate weghts. References Efron, B. (1982). The Jackknfe, the Bootstrap and Other Resamplng Plans. Phladelpha: Socety for Industral and Appled Mathematcs. Hanley, J.A. and McNel, B.J. (1982). The Meanng and Use of the Area under a Recever Operatng Characterstc (ROC) Curve, Radology, 143, Pepe, M.S. (2000). Recever Operatng Characterstc Methodology, Journal of the Amercan Statstcal Assocaton, 95, Rust, K.F. and Rao, J.N.K. (1996). Varance Estmaton for Complex Surveys Usng Replcaton Technques, Statstcal Methods n Medcal Research, 5, SAS Insttute Inc. (1995). Logstc Regresson Examples Usng the SAS System, Verson 6, Frst Edton. Cary, NC: SAS Insttute Inc. SAS Insttute Inc. (1999). SAS/STAT, Verson 8, Chapter 39. Cary, NC: SAS Insttute Inc. Wolter, K.M. (1985). Introducton to Varance Estmaton. New York: Sprnger-Verlag. Acknowledgment We thank Phlp Prmak, Prncpal Programmer at Genzyme Corporaton, for revewng the macros and makng valuable suggestons. Contact Informaton Davd Izrael Abt Assocates Inc. 55 Wheeler St. Cambrdge, MA davd_zrael@abtassoc.com 6

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

Simulation: Solving Dynamic Models ABE 5646 Week 11 Chapter 2, Spring 2010

Simulation: Solving Dynamic Models ABE 5646 Week 11 Chapter 2, Spring 2010 Smulaton: Solvng Dynamc Models ABE 5646 Week Chapter 2, Sprng 200 Week Descrpton Readng Materal Mar 5- Mar 9 Evaluatng [Crop] Models Comparng a model wth data - Graphcal, errors - Measures of agreement

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

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

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

Data Mining: Model Evaluation

Data Mining: Model Evaluation Data Mnng: Model Evaluaton Aprl 16, 2013 1 Issues: Evaluatng Classfcaton Methods Accurac classfer accurac: predctng class label predctor accurac: guessng value of predcted attrbutes Speed tme to construct

More information

Wishing you all a Total Quality New Year!

Wishing you all a Total Quality New Year! Total Qualty Management and Sx Sgma Post Graduate Program 214-15 Sesson 4 Vnay Kumar Kalakband Assstant Professor Operatons & Systems Area 1 Wshng you all a Total Qualty New Year! Hope you acheve Sx sgma

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

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

Mathematics 256 a course in differential equations for engineering students

Mathematics 256 a course in differential equations for engineering students Mathematcs 56 a course n dfferental equatons for engneerng students Chapter 5. More effcent methods of numercal soluton Euler s method s qute neffcent. Because the error s essentally proportonal to the

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

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

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

Life Tables (Times) Summary. Sample StatFolio: lifetable times.sgp

Life Tables (Times) Summary. Sample StatFolio: lifetable times.sgp Lfe Tables (Tmes) Summary... 1 Data Input... 2 Analyss Summary... 3 Survval Functon... 5 Log Survval Functon... 6 Cumulatve Hazard Functon... 7 Percentles... 7 Group Comparsons... 8 Summary The Lfe Tables

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

Estimating Regression Coefficients using Weighted Bootstrap with Probability

Estimating Regression Coefficients using Weighted Bootstrap with Probability Norazan M R, Habshah Md, A H M R Imon Estmatng Regresson Coeffcents usng Weghted Bootstrap wth Probablty NORAZAN M R, HABSHAH MIDI AND A H M R IMON Faculty of Computer and Mathematcal Scences, Unversty

More information

Term Weighting Classification System Using the Chi-square Statistic for the Classification Subtask at NTCIR-6 Patent Retrieval Task

Term Weighting Classification System Using the Chi-square Statistic for the Classification Subtask at NTCIR-6 Patent Retrieval Task Proceedngs of NTCIR-6 Workshop Meetng, May 15-18, 2007, Tokyo, Japan Term Weghtng Classfcaton System Usng the Ch-square Statstc for the Classfcaton Subtask at NTCIR-6 Patent Retreval Task Kotaro Hashmoto

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

Hermite Splines in Lie Groups as Products of Geodesics

Hermite Splines in Lie Groups as Products of Geodesics Hermte Splnes n Le Groups as Products of Geodescs Ethan Eade Updated May 28, 2017 1 Introducton 1.1 Goal Ths document defnes a curve n the Le group G parametrzed by tme and by structural parameters n the

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

Programming in Fortran 90 : 2017/2018

Programming in Fortran 90 : 2017/2018 Programmng n Fortran 90 : 2017/2018 Programmng n Fortran 90 : 2017/2018 Exercse 1 : Evaluaton of functon dependng on nput Wrte a program who evaluate the functon f (x,y) for any two user specfed values

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

Problem Set 3 Solutions

Problem Set 3 Solutions Introducton to Algorthms October 4, 2002 Massachusetts Insttute of Technology 6046J/18410J Professors Erk Demane and Shaf Goldwasser Handout 14 Problem Set 3 Solutons (Exercses were not to be turned n,

More information

Compiler Design. Spring Register Allocation. Sample Exercises and Solutions. Prof. Pedro C. Diniz

Compiler Design. Spring Register Allocation. Sample Exercises and Solutions. Prof. Pedro C. Diniz Compler Desgn Sprng 2014 Regster Allocaton Sample Exercses and Solutons Prof. Pedro C. Dnz USC / Informaton Scences Insttute 4676 Admralty Way, Sute 1001 Marna del Rey, Calforna 90292 pedro@s.edu Regster

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

This module is part of the. Memobust Handbook. on Methodology of Modern Business Statistics

This module is part of the. Memobust Handbook. on Methodology of Modern Business Statistics Ths module s part of the Memobust Handbook on Methodology of Modern Busness Statstcs 26 March 2014 Theme: Donor Imputaton Contents General secton... 3 1. Summary... 3 2. General descrpton... 3 2.1 Introducton

More information

Review of approximation techniques

Review of approximation techniques CHAPTER 2 Revew of appromaton technques 2. Introducton Optmzaton problems n engneerng desgn are characterzed by the followng assocated features: the objectve functon and constrants are mplct functons evaluated

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

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

A Unified Framework for Semantics and Feature Based Relevance Feedback in Image Retrieval Systems

A Unified Framework for Semantics and Feature Based Relevance Feedback in Image Retrieval Systems A Unfed Framework for Semantcs and Feature Based Relevance Feedback n Image Retreval Systems Ye Lu *, Chunhu Hu 2, Xngquan Zhu 3*, HongJang Zhang 2, Qang Yang * School of Computng Scence Smon Fraser Unversty

More information

Variance estimation in EU-SILC survey

Variance estimation in EU-SILC survey Varance estmaton n EU-SILC survey Mārtņš Lberts* * Mathematcal Support Dvson, Central Statstcal Bureau of Latva (Martns.Lberts@csb.gov.lv) bstract: The mplementaton of varance estmaton methods resamplng

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

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

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

A Binarization Algorithm specialized on Document Images and Photos

A Binarization Algorithm specialized on Document Images and Photos A Bnarzaton Algorthm specalzed on Document mages and Photos Ergna Kavalleratou Dept. of nformaton and Communcaton Systems Engneerng Unversty of the Aegean kavalleratou@aegean.gr Abstract n ths paper, a

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

FEATURE EXTRACTION. Dr. K.Vijayarekha. Associate Dean School of Electrical and Electronics Engineering SASTRA University, Thanjavur

FEATURE EXTRACTION. Dr. K.Vijayarekha. Associate Dean School of Electrical and Electronics Engineering SASTRA University, Thanjavur FEATURE EXTRACTION Dr. K.Vjayarekha Assocate Dean School of Electrcal and Electroncs Engneerng SASTRA Unversty, Thanjavur613 41 Jont Intatve of IITs and IISc Funded by MHRD Page 1 of 8 Table of Contents

More information

TN348: Openlab Module - Colocalization

TN348: Openlab Module - Colocalization TN348: Openlab Module - Colocalzaton Topc The Colocalzaton module provdes the faclty to vsualze and quantfy colocalzaton between pars of mages. The Colocalzaton wndow contans a prevew of the two mages

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

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

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

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

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

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

A New Approach For the Ranking of Fuzzy Sets With Different Heights

A New Approach For the Ranking of Fuzzy Sets With Different Heights New pproach For the ankng of Fuzzy Sets Wth Dfferent Heghts Pushpnder Sngh School of Mathematcs Computer pplcatons Thapar Unversty, Patala-7 00 Inda pushpndersnl@gmalcom STCT ankng of fuzzy sets plays

More information

R s s f. m y s. SPH3UW Unit 7.3 Spherical Concave Mirrors Page 1 of 12. Notes

R s s f. m y s. SPH3UW Unit 7.3 Spherical Concave Mirrors Page 1 of 12. Notes SPH3UW Unt 7.3 Sphercal Concave Mrrors Page 1 of 1 Notes Physcs Tool box Concave Mrror If the reflectng surface takes place on the nner surface of the sphercal shape so that the centre of the mrror bulges

More information

Problem Definitions and Evaluation Criteria for Computational Expensive Optimization

Problem Definitions and Evaluation Criteria for Computational Expensive Optimization Problem efntons and Evaluaton Crtera for Computatonal Expensve Optmzaton B. Lu 1, Q. Chen and Q. Zhang 3, J. J. Lang 4, P. N. Suganthan, B. Y. Qu 6 1 epartment of Computng, Glyndwr Unversty, UK Faclty

More information

Some Advanced SPC Tools 1. Cumulative Sum Control (Cusum) Chart For the data shown in Table 9-1, the x chart can be generated.

Some Advanced SPC Tools 1. Cumulative Sum Control (Cusum) Chart For the data shown in Table 9-1, the x chart can be generated. Some Advanced SP Tools 1. umulatve Sum ontrol (usum) hart For the data shown n Table 9-1, the x chart can be generated. However, the shft taken place at sample #21 s not apparent. 92 For ths set samples,

More information

A Simple and Efficient Goal Programming Model for Computing of Fuzzy Linear Regression Parameters with Considering Outliers

A Simple and Efficient Goal Programming Model for Computing of Fuzzy Linear Regression Parameters with Considering Outliers 62626262621 Journal of Uncertan Systems Vol.5, No.1, pp.62-71, 211 Onlne at: www.us.org.u A Smple and Effcent Goal Programmng Model for Computng of Fuzzy Lnear Regresson Parameters wth Consderng Outlers

More information

5.0 Quality Assurance

5.0 Quality Assurance 5.0 Dr. Fred Omega Garces Analytcal Chemstry 25 Natural Scence, Mramar College Bascs of s what we do to get the rght answer for our purpose QA s planned and refers to planned and systematc producton processes

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

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

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

Complex Numbers. Now we also saw that if a and b were both positive then ab = a b. For a second let s forget that restriction and do the following.

Complex Numbers. Now we also saw that if a and b were both positive then ab = a b. For a second let s forget that restriction and do the following. Complex Numbers The last topc n ths secton s not really related to most of what we ve done n ths chapter, although t s somewhat related to the radcals secton as we wll see. We also won t need the materal

More information

Assignment # 2. Farrukh Jabeen Algorithms 510 Assignment #2 Due Date: June 15, 2009.

Assignment # 2. Farrukh Jabeen Algorithms 510 Assignment #2 Due Date: June 15, 2009. Farrukh Jabeen Algorthms 51 Assgnment #2 Due Date: June 15, 29. Assgnment # 2 Chapter 3 Dscrete Fourer Transforms Implement the FFT for the DFT. Descrbed n sectons 3.1 and 3.2. Delverables: 1. Concse descrpton

More information

An Image Fusion Approach Based on Segmentation Region

An Image Fusion Approach Based on Segmentation Region Rong Wang, L-Qun Gao, Shu Yang, Yu-Hua Cha, and Yan-Chun Lu An Image Fuson Approach Based On Segmentaton Regon An Image Fuson Approach Based on Segmentaton Regon Rong Wang, L-Qun Gao, Shu Yang 3, Yu-Hua

More information

Harvard University CS 101 Fall 2005, Shimon Schocken. Assembler. Elements of Computing Systems 1 Assembler (Ch. 6)

Harvard University CS 101 Fall 2005, Shimon Schocken. Assembler. Elements of Computing Systems 1 Assembler (Ch. 6) Harvard Unversty CS 101 Fall 2005, Shmon Schocken Assembler Elements of Computng Systems 1 Assembler (Ch. 6) Why care about assemblers? Because Assemblers employ some nfty trcks Assemblers are the frst

More information

APPLICATION OF MULTIVARIATE LOSS FUNCTION FOR ASSESSMENT OF THE QUALITY OF TECHNOLOGICAL PROCESS MANAGEMENT

APPLICATION OF MULTIVARIATE LOSS FUNCTION FOR ASSESSMENT OF THE QUALITY OF TECHNOLOGICAL PROCESS MANAGEMENT 3. - 5. 5., Brno, Czech Republc, EU APPLICATION OF MULTIVARIATE LOSS FUNCTION FOR ASSESSMENT OF THE QUALITY OF TECHNOLOGICAL PROCESS MANAGEMENT Abstract Josef TOŠENOVSKÝ ) Lenka MONSPORTOVÁ ) Flp TOŠENOVSKÝ

More information

Content Based Image Retrieval Using 2-D Discrete Wavelet with Texture Feature with Different Classifiers

Content Based Image Retrieval Using 2-D Discrete Wavelet with Texture Feature with Different Classifiers IOSR Journal of Electroncs and Communcaton Engneerng (IOSR-JECE) e-issn: 78-834,p- ISSN: 78-8735.Volume 9, Issue, Ver. IV (Mar - Apr. 04), PP 0-07 Content Based Image Retreval Usng -D Dscrete Wavelet wth

More information

7/12/2016. GROUP ANALYSIS Martin M. Monti UCLA Psychology AGGREGATING MULTIPLE SUBJECTS VARIANCE AT THE GROUP LEVEL

7/12/2016. GROUP ANALYSIS Martin M. Monti UCLA Psychology AGGREGATING MULTIPLE SUBJECTS VARIANCE AT THE GROUP LEVEL GROUP ANALYSIS Martn M. Mont UCLA Psychology NITP AGGREGATING MULTIPLE SUBJECTS When we conduct mult-subject analyss we are tryng to understand whether an effect s sgnfcant across a group of people. Whether

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

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

Econometrics 2. Panel Data Methods. Advanced Panel Data Methods I

Econometrics 2. Panel Data Methods. Advanced Panel Data Methods I Panel Data Methods Econometrcs 2 Advanced Panel Data Methods I Last tme: Panel data concepts and the two-perod case (13.3-4) Unobserved effects model: Tme-nvarant and dosyncratc effects Omted varables

More information

ECONOMICS 452* -- Stata 12 Tutorial 6. Stata 12 Tutorial 6. TOPIC: Representing Multi-Category Categorical Variables with Dummy Variable Regressors

ECONOMICS 452* -- Stata 12 Tutorial 6. Stata 12 Tutorial 6. TOPIC: Representing Multi-Category Categorical Variables with Dummy Variable Regressors ECONOMICS 45* -- Stata 1 Tutoral 6 Stata 1 Tutoral 6 TOPIC: Representng Mult-Category Categorcal Varables wth Dummy Varable Regressors DATA: wage1_econ45.dta (a Stata-format dataset) TASKS: Stata 1 Tutoral

More information

ECONOMICS 452* -- Stata 11 Tutorial 6. Stata 11 Tutorial 6. TOPIC: Representing Multi-Category Categorical Variables with Dummy Variable Regressors

ECONOMICS 452* -- Stata 11 Tutorial 6. Stata 11 Tutorial 6. TOPIC: Representing Multi-Category Categorical Variables with Dummy Variable Regressors ECONOMICS * -- Stata 11 Tutoral Stata 11 Tutoral TOPIC: Representng Mult-Category Categorcal Varables wth Dummy Varable Regressors DATA: wage1_econ.dta (a Stata-format dataset) TASKS: Stata 11 Tutoral

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

Skew Angle Estimation and Correction of Hand Written, Textual and Large areas of Non-Textual Document Images: A Novel Approach

Skew Angle Estimation and Correction of Hand Written, Textual and Large areas of Non-Textual Document Images: A Novel Approach Angle Estmaton and Correcton of Hand Wrtten, Textual and Large areas of Non-Textual Document Images: A Novel Approach D.R.Ramesh Babu Pyush M Kumat Mahesh D Dhannawat PES Insttute of Technology Research

More information

Improved Methods for Lithography Model Calibration

Improved Methods for Lithography Model Calibration Improved Methods for Lthography Model Calbraton Chrs Mack www.lthoguru.com, Austn, Texas Abstract Lthography models, ncludng rgorous frst prncple models and fast approxmate models used for OPC, requre

More information

Steps for Computing the Dissimilarity, Entropy, Herfindahl-Hirschman and. Accessibility (Gravity with Competition) Indices

Steps for Computing the Dissimilarity, Entropy, Herfindahl-Hirschman and. Accessibility (Gravity with Competition) Indices Steps for Computng the Dssmlarty, Entropy, Herfndahl-Hrschman and Accessblty (Gravty wth Competton) Indces I. Dssmlarty Index Measurement: The followng formula can be used to measure the evenness between

More information

An Application of the Dulmage-Mendelsohn Decomposition to Sparse Null Space Bases of Full Row Rank Matrices

An Application of the Dulmage-Mendelsohn Decomposition to Sparse Null Space Bases of Full Row Rank Matrices Internatonal Mathematcal Forum, Vol 7, 2012, no 52, 2549-2554 An Applcaton of the Dulmage-Mendelsohn Decomposton to Sparse Null Space Bases of Full Row Rank Matrces Mostafa Khorramzadeh Department of Mathematcal

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

Nonlinear Mixed Model Methods and Prediction Procedures Demonstrated on a Volume-Age Model

Nonlinear Mixed Model Methods and Prediction Procedures Demonstrated on a Volume-Age Model Nonlnear Mxed Model Methods and Predcton Procedures Demonstrated on a Volume-Age Model Prepared by: Shongmng Huang, Shawn X. Meng, Yuqng Yang August, 3 Techncal Report Pub. No.: T/6(3) ISBN No. 978--46-755-3

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

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

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

Course Introduction. Algorithm 8/31/2017. COSC 320 Advanced Data Structures and Algorithms. COSC 320 Advanced Data Structures and Algorithms

Course Introduction. Algorithm 8/31/2017. COSC 320 Advanced Data Structures and Algorithms. COSC 320 Advanced Data Structures and Algorithms Course Introducton Course Topcs Exams, abs, Proects A quc loo at a few algorthms 1 Advanced Data Structures and Algorthms Descrpton: We are gong to dscuss algorthm complexty analyss, algorthm desgn technques

More information

Backpropagation: In Search of Performance Parameters

Backpropagation: In Search of Performance Parameters Bacpropagaton: In Search of Performance Parameters ANIL KUMAR ENUMULAPALLY, LINGGUO BU, and KHOSROW KAIKHAH, Ph.D. Computer Scence Department Texas State Unversty-San Marcos San Marcos, TX-78666 USA ae049@txstate.edu,

More information

A fault tree analysis strategy using binary decision diagrams

A fault tree analysis strategy using binary decision diagrams Loughborough Unversty Insttutonal Repostory A fault tree analyss strategy usng bnary decson dagrams Ths tem was submtted to Loughborough Unversty's Insttutonal Repostory by the/an author. Addtonal Informaton:

More information

Related-Mode Attacks on CTR Encryption Mode

Related-Mode Attacks on CTR Encryption Mode Internatonal Journal of Network Securty, Vol.4, No.3, PP.282 287, May 2007 282 Related-Mode Attacks on CTR Encrypton Mode Dayn Wang, Dongda Ln, and Wenlng Wu (Correspondng author: Dayn Wang) Key Laboratory

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

Performance Evaluation of Information Retrieval Systems

Performance Evaluation of Information Retrieval Systems Why System Evaluaton? Performance Evaluaton of Informaton Retreval Systems Many sldes n ths secton are adapted from Prof. Joydeep Ghosh (UT ECE) who n turn adapted them from Prof. Dk Lee (Unv. of Scence

More information

Finite Population Small Area Interval Estimation

Finite Population Small Area Interval Estimation Journal of Offcal Statstcs, Vol. 23, No. 2, 2007, pp. 223 237 Fnte Populaton Small Area Interval Estmaton L-Chun Zhang 1 Small area nterval estmaton s consdered for a fnte populaton, where the small area

More information

Dealing with small samples and dimensionality issues in data envelopment analysis

Dealing with small samples and dimensionality issues in data envelopment analysis MPRA Munch Personal RePEc Archve Dealng wth small samples and dmensonalty ssues n data envelopment analyss Panagots Zervopoulos Unversty of Western Greece 5. February 2012 Onlne at http://mpra.ub.un-muenchen.de/39226/

More information

A Post Randomization Framework for Privacy-Preserving Bayesian. Network Parameter Learning

A Post Randomization Framework for Privacy-Preserving Bayesian. Network Parameter Learning A Post Randomzaton Framework for Prvacy-Preservng Bayesan Network Parameter Learnng JIANJIE MA K.SIVAKUMAR School Electrcal Engneerng and Computer Scence, Washngton State Unversty Pullman, WA. 9964-75

More information

An Improved Image Segmentation Algorithm Based on the Otsu Method

An Improved Image Segmentation Algorithm Based on the Otsu Method 3th ACIS Internatonal Conference on Software Engneerng, Artfcal Intellgence, Networkng arallel/dstrbuted Computng An Improved Image Segmentaton Algorthm Based on the Otsu Method Mengxng Huang, enjao Yu,

More information

An Optimal Algorithm for Prufer Codes *

An Optimal Algorithm for Prufer Codes * J. Software Engneerng & Applcatons, 2009, 2: 111-115 do:10.4236/jsea.2009.22016 Publshed Onlne July 2009 (www.scrp.org/journal/jsea) An Optmal Algorthm for Prufer Codes * Xaodong Wang 1, 2, Le Wang 3,

More information

Bayesian inference for sample surveys

Bayesian inference for sample surveys Bayesan nference for sample surveys Roderck Lttle Module 3: Bayesan models for complex sample desgns Modelng sample selecton Role of sample desgn n model-based (Bayesan) nference Key to understandng the

More information

(1) The control processes are too complex to analyze by conventional quantitative techniques.

(1) The control processes are too complex to analyze by conventional quantitative techniques. Chapter 0 Fuzzy Control and Fuzzy Expert Systems The fuzzy logc controller (FLC) s ntroduced n ths chapter. After ntroducng the archtecture of the FLC, we study ts components step by step and suggest a

More information

TECHNIQUE OF FORMATION HOMOGENEOUS SAMPLE SAME OBJECTS. Muradaliyev A.Z.

TECHNIQUE OF FORMATION HOMOGENEOUS SAMPLE SAME OBJECTS. Muradaliyev A.Z. TECHNIQUE OF FORMATION HOMOGENEOUS SAMPLE SAME OBJECTS Muradalyev AZ Azerbajan Scentfc-Research and Desgn-Prospectng Insttute of Energetc AZ1012, Ave HZardab-94 E-mal:aydn_murad@yahoocom Importance of

More information

Solutions to Programming Assignment Five Interpolation and Numerical Differentiation

Solutions to Programming Assignment Five Interpolation and Numerical Differentiation College of Engneerng and Coputer Scence Mechancal Engneerng Departent Mechancal Engneerng 309 Nuercal Analyss of Engneerng Systes Sprng 04 Nuber: 537 Instructor: Larry Caretto Solutons to Prograng Assgnent

More information

EECS 730 Introduction to Bioinformatics Sequence Alignment. Luke Huan Electrical Engineering and Computer Science

EECS 730 Introduction to Bioinformatics Sequence Alignment. Luke Huan Electrical Engineering and Computer Science EECS 730 Introducton to Bonformatcs Sequence Algnment Luke Huan Electrcal Engneerng and Computer Scence http://people.eecs.ku.edu/~huan/ HMM Π s a set of states Transton Probabltes a kl Pr( l 1 k Probablty

More information

Learning-Based Top-N Selection Query Evaluation over Relational Databases

Learning-Based Top-N Selection Query Evaluation over Relational Databases Learnng-Based Top-N Selecton Query Evaluaton over Relatonal Databases Lang Zhu *, Wey Meng ** * School of Mathematcs and Computer Scence, Hebe Unversty, Baodng, Hebe 071002, Chna, zhu@mal.hbu.edu.cn **

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

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

Assembler. Shimon Schocken. Spring Elements of Computing Systems 1 Assembler (Ch. 6) Compiler. abstract interface.

Assembler. Shimon Schocken. Spring Elements of Computing Systems 1 Assembler (Ch. 6) Compiler. abstract interface. IDC Herzlya Shmon Schocken Assembler Shmon Schocken Sprng 2005 Elements of Computng Systems 1 Assembler (Ch. 6) Where we are at: Human Thought Abstract desgn Chapters 9, 12 abstract nterface H.L. Language

More information

CSE 326: Data Structures Quicksort Comparison Sorting Bound

CSE 326: Data Structures Quicksort Comparison Sorting Bound CSE 326: Data Structures Qucksort Comparson Sortng Bound Steve Setz Wnter 2009 Qucksort Qucksort uses a dvde and conquer strategy, but does not requre the O(N) extra space that MergeSort does. Here s the

More information

11. HARMS How To: CSV Import

11. HARMS How To: CSV Import and Rsk System 11. How To: CSV Import Preparng the spreadsheet for CSV Import Refer to the spreadsheet template to ad algnng spreadsheet columns wth Data Felds. The spreadsheet s shown n the Appendx, an

More information

Estimating Flow Distributions from Sampled Flow Statistics

Estimating Flow Distributions from Sampled Flow Statistics Estmatng Flow Dstrbutons from Sampled Flow Statstcs Nck Duffeld AT&T Labs Research 80 Park Avenue Florham Park, NJ 07932, USA duffeld@research.att.com Carsten Lund AT&T Labs Research 80 Park Avenue Florham

More information

Query Clustering Using a Hybrid Query Similarity Measure

Query Clustering Using a Hybrid Query Similarity Measure Query clusterng usng a hybrd query smlarty measure Fu. L., Goh, D.H., & Foo, S. (2004). WSEAS Transacton on Computers, 3(3), 700-705. Query Clusterng Usng a Hybrd Query Smlarty Measure Ln Fu, Don Hoe-Lan

More information