Resampling is taking a sample from a sample. Common uses are jackknifing (taking a subsample, eg all values but 1) & bootstrapping (sampling w/ replacement). These techniques can provide a robust estimate of a sampling distribution when it would be difficult or impossible to derive analytically.

learn more… | top users | synonyms

0
votes
0answers
12 views

Evaluate “sampling error” due to lack of data

I'm working on a problem of evaluating the long term success of some investments in some companies around Europe. My problem is that I have the information, before the investment was done, of say 100 ...
0
votes
0answers
6 views

When should the Pasting ensemble method be used instead of Bagging?

Pasting and Bagging are very similar, the main difference being that Bagging samples with replacement (which is called "bootstrapping") while Pasting samples without replacement. I am guessing that ...
0
votes
0answers
11 views

What is the use of BaggingClassifier's bootstrap_features parameter in Scikit-Learn?

Scikit-Learn's BaggingClassifier class has a hyperparameter called bootstrap_features. If I understand correctly, when it is ...
0
votes
1answer
38 views

Cross validation function from which programming language is more appropriate? [closed]

I'd like to use resample to achieve stable results of an unsupervised algorithm that finds clusters in data. I'll use k-fold cross validation repeated many times but I'm in doubt if I should use R or ...
0
votes
0answers
7 views

Error estimation from resampling

In my data analysis I must always keep the errors (or uncertainties). In physics, a result without an error estimate is worthless. Take this trivial example: I measure $x$ and convert it to the ...
0
votes
0answers
55 views

Level of Mutual Information Among Bootstrap Samples vs. Delete-50% Jackknife Samples

Q: I wondered if anyone could offer a mathematical proof or similar, that, on average, delete-50% jackknife samples are not inherently more “independent,” in terms of information content, than ...
0
votes
1answer
45 views

Subsampling to determine a standard error, how does it work?

I need to calculate the standard error on a complicated dataset (> 1700 records) which uses genetic matching. Using bootstrap results in very high computation time (because of the genetic matching). ...
3
votes
1answer
61 views

Resampling pre - post data by shuffling pre/post

Is the following approach a valid way to use resampling to determine if observed changes in pre/post test are significant: The null hypothesis is that there is no difference between the pre and post ...
1
vote
1answer
40 views

Compute confidence interval for univariate Kernel Density Estimator

I've got a univariate dataset (timeseries) for two kind of simulated systems, and I want to explore the differences between the two. To do that, I can build a univariate gaussian KDE for each dataset ...
1
vote
1answer
30 views

Can you acquaint a randomization test with any test?

I want to determine whether my randomizer works properly in a statistical test. One method to do this is a randomization test where we try all possible combinations for two groups. For example: could ...
0
votes
1answer
52 views

How do I show that the two methods of permutation test are both valid?

My main objective is to show the methods described below are really the same. However, I am having difficult both formulating the idea clearly and proving my assertion. Below is my attempt. Suppose ...
1
vote
0answers
24 views

log transform vs. resampling

I want to run a simple regression predicting score on some task, from number of minutes spent doing another activity. My N is ~800. The score variable is normally distributed and measured in ...
2
votes
1answer
82 views

How to Sample from a Randomisation Distribution?

I will use an example to explain my problem, but I think that it can be generalised to any randomization testing that does not allow one to create the null distribution by enumerating of all the ...
0
votes
0answers
28 views

Distribution of shuffled elements in permutation test

I've tried to find an answer to this problem by searching online, but I couldn't, so please forgive me if this question has been asked and answered. :) I have two vectors of length ...
6
votes
1answer
233 views

How to generate a more accurate distribution of sample of random variates?

I want to generate a sample of random variates choosing 5 out of 7 days of the week, $X_1,\ldots,X_5$, such that the aggregate counts resemble a given day-of-week profile, namely $$\mathbb{E}[\mathbb{...
0
votes
0answers
24 views

Modeling rare events

I am looking to model fraudulent cases using logistic regression. However there are tow different datasets which are available. I used to build my model on; it had 4% of fraud cases. My model on ...
0
votes
1answer
59 views

In R package caret, how is linear regression model trained by using resampling?

Resampling is usually used to find the best tuning parameters for a model. However, for some models, such as linear regression model, there is no tuning parameters. In this case, what can we get from ...
1
vote
0answers
19 views

Quantum computing and resampling techniques

Maybe I miss interpreted how does quantum computing work. If I understood well it would allow to perform extreme parallelization by making using a single qubit to perform many calculations at the ...
1
vote
1answer
31 views

Sample Representing a Different Population

I have two sets of populations: containing 1.5 million and 5.5 million units. I need to select a sample out of 5.5 million population so that the sample represents the 1.5 million population based on ...
2
votes
1answer
157 views

Difference between bootstrap and resampling

I am using biological / microarray data. For example, one of my datasets has 50 samples, and 1000 gene attributes. They have 2 labels, Normal and ...
1
vote
1answer
146 views

Cross-validation vs random sampling for classification test

I usually have used cross-validation for testing classification performance. However, I read about the article that random sampling (bootstrapping) works better in many cases. I am not sure which one ...
0
votes
0answers
22 views

optimum bootstrap resamples quantity

If it can be defined, what is the optimum bootstrap resamples quantity? My specific aim is to determine confidence level (CL). notes: I am not a statistician. I want to find CL for failure rates of ...
1
vote
1answer
49 views

Confused about the sample size in large sample confidential interval and population proportion intertal

My professor said when sample size is large, it will has the following formula, but how big is the sample can be considered to be a large sample and use the following formula? I did some questions ...
1
vote
1answer
39 views

Difference between monte carlo based analysis and a hypothesis test

Is it reasonable to use Monte Carlo methods to resample a dataset of weekly rainfall amounts to statistically test for difference between two timeseries? That is, randomly pull ~30 paired observations ...
0
votes
1answer
35 views

Applying SMOTE and PCA to high dimensional data giving low accuracy

I have a high dimensional datasets of around 2300+ columns. The dataset consist of two class labels of which one is extremely biased and occurs less than 10%. I looked at the various algorithms and ...
0
votes
0answers
27 views

Would splitting an unbalanced data into training and test data help for better classifier accuracy?

This will be my first post here. I have a not-so unbalanced data set with a 3 class variable having a distribution of approximately 48%,35% and 17% for the three of them. Is it advisable to preprocess ...
0
votes
0answers
36 views

Artificially adding more points x, y in linear regression

If I have $x$, $y$ data pairs and I am fitting $y = ax + b$, so I am using a standard OLS model. My questions are: If I take a particular value of the tuple (x,y), say $x_5$, $y_5$ pair and add it ...
0
votes
0answers
28 views

Creating a bootstrap null (H0) distribution. Investigating alternatives

I'm investigating the possibility of creating a non parametric bootstrap null distribution for hypothesis testing in multivariable regression analysis. The null hypothesis is based on the absence of ...
0
votes
1answer
53 views

How do I bootstrap with R correctly without increasing the degree of freedom?

I need to compare survey responses from my experimental (referred as "FS" below) and control group (referred as "GV" below) using independent samples t-test. As the two groups have unequal sample ...
0
votes
1answer
64 views

Resampling under the null versus the alternative hypothesis

I'm looking at a community of organisms using simultaneous GLMs via the mvabund package. The manyglm function from this package ...
0
votes
0answers
8 views

Sampling from a probability list/set efficiently [duplicate]

Let $A = ((a_1, p_1), (a_2, p_2), \dots)$ be a list of elements $a$ which associated probabilities $p$ where $\sum p_i = 1$. $A$ can also be a set, as all $a_i$ are distinct and their order doesn't ...
0
votes
0answers
45 views

minority class resampling and classification accuracy inflation

I have a question whether resampling (with synthetic data) to fix uneven class distribution inflates classifier accuracy? I've used SMOTE algorithm to increase the number of instances in 4 minority ...
3
votes
1answer
35 views

Is training error ever “good enough” when we have a large enough sample?

I recently had a (more experienced) coworker tell me that when you have a large enough sample, training error should be "good enough" to assess model performance. His point was that with a ...
1
vote
1answer
102 views

How to re-sample particle filter's particles for a 1D door/wall problem

So assuming your implementation of the motion model and sensor model is at a somewhat satisfying level, the question then is how do I stabilize localization with the re-sampling step. I'm currently ...
0
votes
0answers
16 views

Resampling for comparative purposes

I'm here with a question concerning resampling. My issue: I measured the stature of an individual and found him markedly tall when compared with other eight subject from the same population. However, ...
0
votes
0answers
19 views

runs of events - resampling logic

Let's say I have a possible event (say, machinery failure) that occurs at a certain probability (say, 30% of the time), and batches of trials (say, production runs of length N). Using resampling logic,...
0
votes
0answers
9 views

Is there a term that describes sensitvity to subsampling?

Assume I have some sample $X$ from a distribution and I use it to calculate some statistic $S$. Sometimes one wishes to evaluate how $S$ changes when calculated from subsamples of $X$. Is there a ...
0
votes
0answers
22 views

Coefficient significance for resampling

I have run the same regression on 10000 samples drawn with stratified random sampling, making use of "simulate" in STATA (the drawn samples are smaller than the original dataset). This gives me a new ...
1
vote
0answers
58 views

Standard error of an estimate from a non-probability sample?

Assume one tries to estimate the mean price of an item by employing a non-probability sample due to the lack of a survey frame of the universe of all items. How would I go about calculating the ...
1
vote
1answer
220 views

Bootstrap methodology. Why resample “with replacement” instead of random subsampling?

The bootstrap method has seen a great diffusion in the last years, I also use it a lot, especially because the reasoning behind is quite intuitive. But that's one thing I don't understand. Why Efron ...
2
votes
1answer
37 views

Updating the head or tail of an ordered subset of rows using data.table in R

The concept is simpler than my title. I have a data.table that represents a sample taken from a population. My goal is to test the performance of several different prediction algorithms across sample ...
0
votes
0answers
15 views

Estimating performance of an iterated random algorithm

I have an algorithm that has a certain probability of computing an answer after a randomly distributed runtime. The probability may be less than 1 even for infinite time in a single run of the ...
1
vote
0answers
32 views

Visually Compare Different Sampling Methods [closed]

I am working on a project exploring results from a dataset, when using different sampling designs on that dataset. It is really fun thus far, and I have been learning a great deal of things especially ...
2
votes
0answers
123 views

Estimation on evolving distribution with small updates

I have a set $X$ of $10^6$ elements and a time series of probability distributions $\mu_1,\mu_2,\ldots$ on $X$. I want to estimate the expected value of a function $f$ over each $\mu_t$. It is easy to ...
0
votes
0answers
57 views

Training set selection

I have the following question for a project I'm working on. I am trying to find the best strategy to select the best training set in a dataset. I have a dataset with a few billions rows. I am trying ...
3
votes
1answer
164 views

Representative resampling

I am working with a population in which each individual has, among others, 6 observed variables that can be 0 or 1: $X_i \sim Bernoulli(p_i),\ i=1,...,6$ . I know the "true" value for the ...
2
votes
2answers
55 views

How to empirically show that a certain quantity approximately follows a normal distribution?

To motivate some theoretical work, I need to show that two certain variables (say $X$ and $Y$) approximately follow a normal distribution in actual datasets. I have one large dataset with about $300$ ...
0
votes
0answers
139 views

Resampling for unbalanced data in cross-validation

Resampling the data prior to classification is one of the techniques dealing with unbalanced dataset. I then consider down-sampling and ...
0
votes
0answers
153 views

Sampling, feature selection and preprocessing in cross validation

To brief my question, I want to clarify the order of parameter tuning and the correctness of the flow in my scheme. In my classification scheme, there are several steps including: SMOTE (Synthetic ...
1
vote
1answer
198 views

Bootstrapping a t-test in R

I have two groups of individuals (22 in each group), which I compared using a t-test. The difference between groups was non-significant (p = .17). Because the p-value was quite low, my supervisor ...