{ From user David, Model Simple_bootstrapping at 23-Jan-2013 2:04:42 PM } SoftwareVersion 4.5.0 { System Variables with non-default values: } SampleSize := 1000 TypeChecking := 1 Checking := 1 SaveOptions := 2 SaveValues := 0 {!40300|Sys_DomainSelfIndex := 1} {!40400|Sys_AllNullTreatment := 1} {!40300|ProactivelyEvaluate Index: 1} Attribute Reference Attribute Date_bough Attribute Notes Attribute Additional Attribute New_attribute AskAttribute Reference,Function,Yes AskAttribute Notes,Function,Yes AskAttribute New_attribute,Function,Yes AskAttribute Reference,Module,Yes AskAttribute Notes,Module,Yes AskAttribute Value,Variable,Yes AskAttribute Reference,Variable,Yes AskAttribute Notes,Variable,Yes Model Simple_bootstrapping Title: Simple Bootstrapping Description: Bootstrapping is a technique for estimating the confidence in a statistical estimator. This model demonstrates the technique with a kurtosis estimator, but the technique is the same for any estimator, be it as simple as variance, or as complex as a coefficient obtained from a complex data fitting process, or even the results of an importance analysis (rank correlation).~ ~ Given a data set, the Kurtosis function computes the sample kurtosis. Of course, because the sample is finite sized, this is only an estimate of the true kurtosis of the underlying variable. The question of interest is how much sampling error is there in the estimated value? ~ ~ The bootstrapping technique estimates this by resampling the original data with replacement, to obtain a resampled data set. We then compute the same statistic using the resampled data set. By repeating with different random re-samplings, a distribution on the estimated statistic is obtained. When bootstrapping is employed, it is usually assumed that this bootstrapped distribution reflects the actual sampling error. ( Which is, of course, not always the case. Bootstrapping often underestimates the true uncertainty). Author: Lonnie Chrisman, Ph.D. Date: Thu, Nov 15, 2007 3:18 PM SaveAuthor: David SaveDate: Wed, Jan 23, 2013 2:04 PM DefaultSize: 48,24 DiagState: 2,44,27,663,300,17 WindState: 2,156,22,507,576 FontStyle: Arial, 15 FileInfo: 0,Model Simple_bootstrapping,2,2,0,0,C:\Documents and Settings\David\Desktop\work\Changes to wiki\13291 - 1.ana {!40400|Att_clearTypeFonts: -1} Index Data_index {!40000|Att_PrevIndexValue: [1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100]} Title: Data Index Definition: 1..100 NodeLocation: 88,64,1 NodeSize: 48,24 Decision Data Title: Data Definition: Table(Data_index)(~ 46.54689464540134,33.59052476299295,36.30195178986784,18.81388044292626,27.94932104962063,37.42361643962924,31.39774617789041,16.4681303560624,49.34146435364882,58.17883657995901,41.29975279150762,24.53984056992109,40.59873471205015,27.80030741960258,47.72934617827674,36.25369816752684,43.0040291728602,53.32536147206682,39.24572612252306,34.37859998706048,21.68284002821039,24.13441899414581,40.29531655063698,39.63574403254375,67.40088373411106,49.82017975156164,71.55978573536723,44.71726424281,29.48920775536967,46.32169484314797,57.95961538404821,56.02591772758763,28.00573976879835,22.62760175149616,39.361127286537,45.74550815849261,33.50933495348677,46.55467600542058,59.34861987431938,27.00016770557577,63.9568264648268,45.2533468226497,41.1485320223223,31.36165842064762,25.580158769252,39.9126878163783,74.03423262006353,36.61172869012097,43.655938148434,50.130945103632,20.13542792586186,49.44135958855424,66.13497895832472,15.47835143273515,45.23753340895286,41.52862369360324,26.4703215723682,33.4687245172565,31.63686951567698,41.95891060823327,48.70974175259086,23.60263472815565,36.3772918629851,71.00235050183211,52.7051783566595,18.13296520754454,19.21642131052605,29.32236360264665,31.05509755315697,51.01209312571965,34.85957262213314,27.05049374091342,35.85710568511622,32.14934098616448,45.91248849966186,51.73742909740559,24.24421691054571,42.37003585544408,22.4075790752195,42.14914575226092,29.01763469828295,32.49892860190535,28.02726573712399,14.78574281045808,36.88782551022037,46.81324963863711,22.04984161108635,58.99059799728074,63.70371832104469,24.30966737829607,33.56232044881197,53.36885542445143,33.49147929349488,27.46284449067303,52.80019583040598,27.43610246368557,40.03378291554098,51.38405355126895,25.51930072845581,41.25821868575696) NodeLocation: 200,64,1 NodeSize: 48,24 ValueState: 2,740,77,487,327,1,MIDM GraphSetup: {!40000|Att_contlinestyle Graph_primary_valdim:0} Objective Kurtosis_of_data Title: Kurtosis of data Definition: Kurtosis(Data,Data_index) NodeLocation: 344,64,1 NodeSize: 48,24 ValueState: 2,665,94,416,303,0,MIDM Index Bootstrap_index Title: Bootstrap index Definition: copyindex(Data_index) NodeLocation: 88,128,1 NodeSize: 48,24 Chance Resampled_data Title: Resampled data Definition: data[ @Data_index = Random(uniform(1,size(Data_index),integer:true),over:Data_index,Bootstrap_index) ] NodeLocation: 200,128,1 NodeSize: 56,24 ValueState: 2,114,260,588,571,0,MIDM ReformVal: [Bootstrap_index,Data_index] Variable Bootstrapped_kurtosi Title: bootstrapped kurtosis Definition: Kurtosis(Resampled_data,Data_index) NodeLocation: 344,128,1 NodeSize: 56,24 ValueState: 2,227,454,416,303,1,MIDM Objective Statistics_for_kurto Title: Statistics for Kurtosis Estimate Definition: Statistics(Bootstrapped_kurtosi,Bootstrap_index ) NodeLocation: 464,208,1 NodeSize: 56,32 ValueState: 2,628,69,416,303,0,MIDM GraphSetup: Statsselect:[1, 1, 1, 1, 1, 1, 0, 0 ] Objective Distribution_of_kurt Title: Distribution of Kurtosis estimate Definition: Pdf(Bootstrapped_kurtosi,Bootstrap_index,smoothingMethod:1) NodeLocation: 344,208,1 NodeSize: 56,32 ValueState: 2,549,440,632,355,1,MIDM ReformVal: [Sys_LocalIndex('STEP'),DensityIndex] {!40000|Att_CoordinateIndex: DensityIndex} {!-40199|MsgBox("Adding this encrypted module into your model will cause your entire model and any linked modules or libraries to become encrypted next time you save. Once encrypted, encryptioon is irreversible, so you may want to cancel at this time without loading this module.~ ~ Note: Upgrading to Analytica 4.2 or later, where encrypted models can be added without forcing your entire model to be encrypted, will avoid this problem.",33,"Warning: Encryption is irreversible")} OBFUSCATED OHFDPEEDV4.2 @