karthik/Calculations
1
0
Fork 0
Code Issues Pull Requests repo.project_board Releases Wiki Activity

Removed the measurement of autocorrelation, modifications made for the bootstrap sample size and number to be user-defined.

Browse Source
This commit is contained in:
Karthik 2021-08-19 12:43:23 +02:00
parent e29c923723
commit 8ac0a018f2
1 changed files with 19 additions and 36 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

55
+Simulator/@TwoDimensionalMOT/bootstrapErrorEstimation.m
View File

@ -1,39 +1,22 @@
function [LoadingRate, StandardError, ConfidenceInterval] = bootstrapErrorEstimation(this, ovenObj, NumberOfLoadedAtoms) function [LoadingRate, StandardError, ConfidenceInterval] = bootstrapErrorEstimation(this, ovenObj, NumberOfLoadedAtoms)
n = this.NumberOfAtoms; n = this.NumberOfAtoms;
NumberOfTimeSteps = int64(this.SimulationTime/this.TimeStep); SampleLength = this.BootstrapSampleLength;
NumberOfBootsrapSamples = this.BootstrapSampleNumber;
Autocorrelation = autocorr(NumberOfLoadedAtoms,'NumLags', double(NumberOfTimeSteps - 1)); MeanCaptureRatioInEachSample = zeros(1,NumberOfBootsrapSamples);
for SampleNumber = 1:NumberOfBootsrapSamples
if Autocorrelation(1)~=0 BoostrapSample = datasample(NumberOfLoadedAtoms, SampleLength); % Sample with replacement
CorrelationFactor = table(Helper.findAllZeroCrossings(linspace(1, double(NumberOfTimeSteps), double(NumberOfTimeSteps)), Autocorrelation)).Var1(1); MeanCaptureRatioInEachSample(SampleNumber) = mean(BoostrapSample) / n; % Empirical bootstrap distribution of sample means
if ~isnan(CorrelationFactor)
SampleLength = floor(CorrelationFactor);
NumberOfBootsrapSamples = 1000;
MeanCaptureRatioInEachSample = zeros(1,NumberOfBootsrapSamples);
for SampleNumber = 1:NumberOfBootsrapSamples
BoostrapSample = datasample(NumberOfLoadedAtoms, SampleLength); % Sample with replacement
MeanCaptureRatioInEachSample(SampleNumber) = mean(BoostrapSample) / n; % Empirical bootstrap distribution of sample means
end
LoadingRate = mean(MeanCaptureRatioInEachSample) * ovenObj.ReducedFlux;
Variance = 0; % Bootstrap Estimate of Variance
for SampleNumber = 1:NumberOfBootsrapSamples
Variance = Variance + (MeanCaptureRatioInEachSample(SampleNumber) - mean(MeanCaptureRatioInEachSample))^2;
end
StandardError = sqrt((1 / (NumberOfBootsrapSamples-1)) * Variance) * ovenObj.ReducedFlux;
ts = tinv([0.025 0.975],NumberOfBootsrapSamples-1); % T-Score
ConfidenceInterval = LoadingRate + ts*StandardError; % 95% Confidence Intervals
else
LoadingRate = nan;
StandardError = nan;
ConfidenceInterval = [nan nan];
end
else
LoadingRate = nan;
StandardError = nan;
ConfidenceInterval = [nan nan];
end end
LoadingRate = mean(MeanCaptureRatioInEachSample) * ovenObj.ReducedFlux;
Variance = 0; % Bootstrap Estimate of Variance
for SampleNumber = 1:NumberOfBootsrapSamples
Variance = Variance + (MeanCaptureRatioInEachSample(SampleNumber) - mean(MeanCaptureRatioInEachSample))^2;
end
StandardError = sqrt((1 / (NumberOfBootsrapSamples-1)) * Variance) * ovenObj.ReducedFlux;
ts = tinv([0.025 0.975],NumberOfBootsrapSamples-1); % T-Score
ConfidenceInterval = LoadingRate + ts*StandardError; % 95% Confidence Intervals
end end