{ From user Lonnie, Model Simple_bootstrapping at Thu, Nov 15, 2007 3:39 PM~~
}
Softwareversion 4.0.1
{ System Variables with non-default values: }
Typechecking := 1
Checking := 1
Saveoptions := 2
Savevalues := 0
Model Simple_bootstrapping
Title: Simple Bootstrapping
Description: Bootstrapping is a technique for estimating the confidenc~~
e in a statistical estimator. This model demonstrates the technique ~~
with a kurtosis estimator, but the technique is the same for any esti~~
mator, be is as simple as variance, or as complex as a coefficient ob~~
tained 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 esti~~
mate of the true kurtosis of the underlying variable. The question o~~
f 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 comp~~
ute the same statistic using the resampled data set. By repeating wi~~
th different random re-samplings, a distribution on the estimated sta~~
tistic is obtained. When bootstrapping is employed, it is usually as~~
sumed that this bootstrapped distribution reflects the actual samplin~~
g 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: Lonnie
Savedate: Thu, Nov 15, 2007 3:39 PM
Defaultsize: 48,24
Diagstate: 1,44,27,550,300,17
Windstate: 2,617,54,507,576
Fontstyle: Arial, 15
Fileinfo: 0,Model Simple_bootstrapping,2,2,0,0,W:\Analytica\ExecDebug\~~
Example Models\Data Analysis\Bootstrapping.ana
Index Data_index
Title: Data Index
Definition: 1..100
Nodelocation: 88,64,1
Nodesize: 48,24
{!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]}
Decision Data
Title: Data
Definition: Table(Data_index)(~
46.54689464540134,33.59052476299295,36.30195178986784,18.813880442926~~
26,27.94932104962063,37.42361643962924,31.39774617789041,16.468130356~~
0624,49.34146435364882,58.17883657995901,41.29975279150762,24.5398405~~
6992109,40.59873471205015,27.80030741960258,47.72934617827674,36.2536~~
9816752684,43.0040291728602,53.32536147206682,39.24572612252306,34.37~~
859998706048,21.68284002821039,24.13441899414581,40.29531655063698,39~~
.63574403254375,67.40088373411106,49.82017975156164,71.55978573536723~~
,44.71726424281,29.48920775536967,46.32169484314797,57.95961538404821~~
,56.02591772758763,28.00573976879835,22.62760175149616,39.36112728653~~
7,45.74550815849261,33.50933495348677,46.55467600542058,59.3486198743~~
1938,27.00016770557577,63.9568264648268,45.2533468226497,41.148532022~~
3223,31.36165842064762,25.580158769252,39.9126878163783,74.0342326200~~
6353,36.61172869012097,43.655938148434,50.130945103632,20.13542792586~~
186,49.44135958855424,66.13497895832472,15.47835143273515,45.23753340~~
895286,41.52862369360324,26.4703215723682,33.4687245172565,31.6368695~~
1567698,41.95891060823327,48.70974175259086,23.60263472815565,36.3772~~
918629851,71.00235050183211,52.7051783566595,18.13296520754454,19.216~~
42131052605,29.32236360264665,31.05509755315697,51.01209312571965,34.~~
85957262213314,27.05049374091342,35.85710568511622,32.14934098616448,~~
45.91248849966186,51.73742909740559,24.24421691054571,42.370035855444~~
08,22.4075790752195,42.14914575226092,29.01763469828295,32.4989286019~~
0535,28.02726573712399,14.78574281045808,36.88782551022037,46.8132496~~
3863711,22.04984161108635,58.99059799728074,63.70371832104469,24.3096~~
6737829607,33.56232044881197,53.36885542445143,33.49147929349488,27.4~~
6284449067303,52.80019583040598,27.43610246368557,40.03378291554098,5~~
1.38405355126895,25.51930072845581,41.25821868575696)
Nodelocation: 200,64,1
Nodesize: 48,24
Valuestate: 2,740,77,416,303,1,MIDM
Graphsetup: {!40000|Att_contlinestyle Graph_primary_valdim:0}
Objective Kurtosis_of_data
Title: Kurtosis of data
Definition: Kurtosis(Data,Data_index)
Nodelocation: 320,64,1
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),inte~~
ger:true),over:Data_index,Bootstrap_index) ]
Nodelocation: 200,128,1
Nodesize: 48,24
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: 320,128,1
Nodesize: 52,24
Valuestate: 2,216,226,416,303,1,MIDM
Objective Statistics_for_kurto
Title: Statistics for Kurtosis Estimate
Definition: Statistics(Bootstrapped_kurtosi,Bootstrap_index )
Nodelocation: 440,193,1
Nodesize: 48,31
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)
Nodelocation: 320,193,1
Nodesize: 48,31
Valuestate: 2,549,108,632,355,1,MIDM
Reformval: [Sys_localindex('STEP'),Densityindex]
{!40000|Att_coordinateindex: Densityindex}
Close Simple_bootstrapping