Questions tagged [bootstrap]

The bootstrap is a resampling method to estimate the sampling distribution of a statistic.

Filter by
Sorted by
Tagged with
4
votes
0answers
37 views

Bootstrap consistency

In the usual setup of a statistical model, suppose that $T_n(X_1,\dots,X_n)$ is a strongly consistent estimator of a parameter $\theta$. Let $X^*_1,\dots,X^*_n$ be a random sample from the empirical ...
0
votes
0answers
6 views

Bootstrap textbook [duplicate]

I used to make good reference to an excellent course on non-parametric statistics but it just got taken down. Maybe its time to find a decent text on bootstrap confidence intervals and smoothing ...
0
votes
0answers
20 views

Bootstrap in Design of Experiments [closed]

Does it make sense to perform bootstrap analysis for coefficients in a Design of Experiments data set, given that excluding treatments can introduce multicollinearity?
0
votes
0answers
22 views

What is wrong with my implementation of bootstrapping? [closed]

My understanding of the bootstrap is as follows. Suppose we have some data $x = (x_1, \dots, x_2)$ that are realizations from a law $\mathcal{P}_{\theta_0}$. We would like to estimate the distribution ...
1
vote
0answers
15 views

Bootstrap iteration as random effect

I have a theoretical question. Would it make sense to replicate a dataset via bootstrapping and then perform a mixed effect analysis using the data from all iterations, specifying the specific ...
0
votes
0answers
21 views

How to bootstrap subsamples in R [closed]

I'm trying to compare 4 groups means (using bootstrap approach) but my groups are unequal in size. To make a better comparison, I would like to set my bootstrap sample sizes to all be the same (i.e. ...
0
votes
0answers
7 views

Testing First-Order Stochastic Dominance of Two Samples from Discrete Distributions

Suppose that $X$ and $Y$ are two independent random variables with common finite support on $\mathcal S =\{0,1,...,K\}$ and let $F_X$ and $F_Y$ be their CDFs. Let $\{X_i\}_{i=1}^n$ and $\{Y_j\}_{j=1}^...
1
vote
0answers
19 views

Different seed gives different LASSO output

So I am using LASSO in order to reduce the amount of independent variables (~150) I have for my logistic model (n=1200). However, when doing so, the end result (i.e. number of predictors) it chooses ...
0
votes
0answers
17 views

Valid significance testing of coefficients when performing model selection

I want to know how I can validly perform permutation (or bootstrap) significance testing of model coefficients when I am also using the same dataset to perform model selection. My model is not the ...
7
votes
4answers
858 views

What is the definition of machine learning (vs classical statistics), and can methods such as MCMC and bootstrapping be considered ML? [duplicate]

I'm in the process of writing about the differences between machine learning and classical statistics. I've been looking for some authoritative sources that would give a good, clear, plain-English ...
1
vote
0answers
16 views

Weighted Block Bootstrap

In this diagram (from 'An introduction to the Bootstrap') the black circles represent the original time series while the white circles represent the sampled bootstrap realization. This bootstrap ...
0
votes
0answers
47 views

comparing two partial correlation coefficients (spearman)

I have carried out the following two spearman partial correlations using the "ppcor" package in R: -- Coefficient 1 = Correlation of A1 with B controlling for C; -- Coefficient 2 = Correlation of A2 ...
1
vote
1answer
89 views

Is it correct to say that the nonparametric BCa (bootstrap confidence intervals) is assumptions-free?

I'd like to use the BCa algorithm to compute 95% confidence limits (maybe let's not call it intervals) for the medians of two population samples. I'd like to visually inspect and conclude whether the ...
0
votes
0answers
34 views

How to bootstrap a logistic regression model? [duplicate]

Currently I have a dataset with 4712 records focused on binary classification. I have divided my data into train(70%) and test (30%) split. I run a gridsearchcv on my train data and choose the best ...
1
vote
0answers
26 views

How to bootstrap the time series data for assets portfolio mean - variance optimization?

As known among professionals and amateur investors, the classical Markovitz portfolio mean - variance optimization yields very unstable results that perform poorly in the real world. One of the ...
0
votes
0answers
10 views

Possible NAs in Beslow estimation with repeated data?

I would like to seek your advice on a problem I encountered recently. I bootstrap on a test that includes a Breslow estimate for survival data. Since these are meta-analyzes, I am bootstrapping on ...
1
vote
0answers
12 views

Count distinct from a bootstrapped sample

I have a dataset of N items, with attribute V which can be repetitive, for example: ...
4
votes
2answers
648 views

Simulating Buffon's Needle in R [closed]

I have the following question I'm trying to write a simulation for: Let T_1 be the number of crossings in n tosses of the needle, then $$E_1 = T_1d/(nl)$$ is an unbiased estimator of 2/π. Write a ...
1
vote
0answers
25 views

How to bootstrap a statistic calculed in a meta-analyis?

I have calculated a statistic in a meta-analysis of individual data and I want to bootstrap it to obtain the variance (no analytic expression is available) to make a test. First, I was naively ...
0
votes
1answer
20 views

Do all non-parametric statistical tests involve bootstrapping, simulation, or permutation test?

All of the non-parametric tests I learned about involved bootstrapping, simulation of large samples of some random variable, or permutation tests. Are there non-parametric tests without simulation or ...
1
vote
1answer
32 views

Testing how well an unweighted linear regression explains variation compared to a weighted linear regression

I have a weighted and an unweighted linear regression (same dataset) and need a measure of how well one model can explain variation relative to the other (i.e. are the R-squared values significantly ...
2
votes
0answers
27 views

Poisson Binomial Distribution - confidence intervals for sum of unequal probabilities estimated with uncertainty

I believe my question is related to, but distinct from this one: Poisson Binomial Distribution - confidence intervals I am working to estimate species richness by summing the results of 12 individual ...
0
votes
0answers
16 views

Difference between cross-validation, backtesting, historical simulation, Monte Carlo simulation, bootstrap replication?

In finance, to determine if a trading strategy is better than others, or to optimize the parameters of a model, the following statistical techniques are often employed, often one over the others ...
0
votes
0answers
3 views

How to build standard error and confidence intervals from Buckland variance estimator after variable selection

In this paper https://www.sciencedirect.com/science/article/pii/S0167947307003957 it is shown how to build an estimator of the variance of estimates computed after variable selection. It evaluates ...
0
votes
1answer
43 views

Standard error of the $n^n$ bootstrap means

I need to show that the standard error of the $n^n$ bootstrap means is $SE^*(\bar{Y^*}) = \frac{S\sqrt{n-1}}{n}$, where $\bar{Y^*}$ is the sample mean of a randomly drawn bootstrap sample, and $S^2 = \...
0
votes
0answers
7 views

Nonparametric bootstrap to construct confidence intervals for Cohen's Kappa Coefficient

I was recently asked to perform nonparametric bootstrap to construct 95% confidence intervals for $\kappa$ using normal approximation, but I'm not sure how to do this. The only data I was given was ...
2
votes
2answers
58 views

What are the uses of bootstrap resampling?

I'm still wrapping my head around bootstrapping but am struggling to think of how it is applied. I have looked at: Explaining to laypeople why bootstrapping works So far, my understanding is that ...
1
vote
1answer
53 views

Common practice for validating classifiers in medical statistics

We are writing a paper, where we suggest a new classifier. We have a data base with about 1400 medical cases. Would it be sufficient just to divide the dataset into training 70% and test 30%? Or ...
1
vote
0answers
12 views

Bootstrap t-test and associated barplots [closed]

I am trying to figure out the code that will run a bootstrap t-test on multivariate data. I have two treatments in this experiment (control and trmt). The response variable is plant cover (there are ...
0
votes
1answer
38 views

How many experiments do I need to perform?

I'm currently collecting time series data from an experiment. I'm collecting the position of an object in 1D, i.e. $x$, at a rate of 60 times per second. I'm then interested in summary statistics. The ...
0
votes
0answers
10 views

Error-bars on split half reliability

Say I administer a test to 30 students. I randomly divide the test questions into two parts and score each half for each student. I then calculate the correlation coefficient between scores on each ...
2
votes
1answer
82 views

Sampling distribution is not normal. How is that possible?

As central limit theorem suggests, sampling distribution is approaching normal on the large sample sizes regardless of the initial distribution of the variable. And it's always been true for me until ...
0
votes
0answers
13 views

Bootstrap hypothesis test of sample difference for a general function f(x)

I have two samples, $x$ and $y$. $x$ is generated from the multinomial $Mult(n_x, \pi_x)$ and y from $Mult(n_y, \pi_y)$. I am interested in the quantities $f(\pi_x)$ and $f(\pi_y)$. Specifically, I ...
1
vote
0answers
21 views

Biased bootstrap: how to construct confidence intervals?

For a project I am trying to evaluate a set of predictors using logistic regression. So I have defined a model selection procedure that I applied on the data, and estimate an AUC to assess model ...
3
votes
2answers
89 views

Calculate accelerated bootstrap interval in R

I am trying to to calculate bootstrap confidence interval on an index calculated from a vector of values, and if the index is significantly greater than 0 in R. For example, the vector of length 6: <...
1
vote
1answer
27 views

Bootstrapped data to fit models and comparison

I have a large data set, say 40k row by 60 columns. Now I do bootstrapping (resample the rows with replacement) on this whole data set for 10k times. Then I used each of the 10k data sets to fit a ...
2
votes
0answers
26 views

Bootstrapping for time series modeling

I am developing a model to decompose a univariate time series dataset into three components: seasonal, trend, and a piecewise semi-linear component with a sawtooth pattern (repeated gradual decline ...
0
votes
0answers
35 views

Bootstrapping for mediation in multilevel model with binary outcome

I am trying to do a mediation analysis with a multilevel model, where the outcome is binary and the predictor and mediator are continous. I wanted to do bootstrapping and then comparison of confident ...
0
votes
1answer
35 views

Question about the use of the word “standard error” in several situations

I'm now reading the book Computer Age Statistical Inference and I'm not sure what the standard error actually refers to in several cases. In Page 9, the author defines for two data sets AML and ALL (...
1
vote
0answers
11 views

Are there any package for stratified bootstrap re-sampling for principal component analysis in R? [closed]

It is possible to bootstrap principal component analysis (see package bootSVD) It is possible to conduct stratified bootstrapping (see package boot, option strata). However, is it possible to conduct ...
0
votes
0answers
29 views

Bootstrapping of data, in R, for improving the homogeneity of variance

I have a data collected on five point likert scale; aims at studying, the level of preference individuals gives to the different factors, (such as Tax Benefits, Maturity period, Rate of Returns ...) ...
1
vote
2answers
55 views

Bootstrapped confidence intervals: different values at every computation?

I am running an ANOVA and computing bootstrapped CIs for my effect sizes. I am using for this the Measures of Effect Size Toolbox for Matlab. To my surprise, if I re-run the same line of code, i.e. ...
1
vote
1answer
35 views

Can't .632+ rule be computed for any kind of outcome and prediction score?

It seems that the R packages I found around for computing the .632+ estimation of prediction error work only with categorical outcomes. Why is that? Looking at the formulas in Efron 1997 paper it ...
0
votes
0answers
21 views

Trend detection using Mann Kendall test and block bootstrap technique

I have to analyze the trend in rainfall time series data. Many literatures suggest that the nonparametric block bootstrap technique incorporating the Mann-Kendall trend test is suitable for trend ...
0
votes
0answers
19 views

How does Data Augmentation work for supervised learning models?

I've ran into a few Kaggle competitions where the winning solution used data augmentation, and a new ML platform, which claimed to help with Data Augmentation. Use cases were imbalanced classes and ...
2
votes
0answers
31 views

Bootstrapping across residuals for heteroscedastic data

I have a highly heteroscedastic 2d dataset (x,y) where both x and y cover about 3 orders of magnitude; therefore, although the % error is roughly constant and normally distributed across x, the ...
0
votes
1answer
71 views

Can we not use Bootstrapping for all small sample ML use cases?

This might be a noob question and excuse me for my naivette here. But we know in Statistics, we can use Bootstrapping to generate more samples repeatedly from the same sample. So on problems where ...
1
vote
0answers
11 views

How to bootstrap residuals of two-way clustered standard errors [closed]

I'm doing a two stage model that I'm trying to bootstrap to get the covariance matrix of the second stage. My data is panel, with observations at the country quarter level. In the first stage I ...
5
votes
0answers
54 views

BCA bootstrap intuition

The papers below claims that BCa bootstrap improves bootstrap estimate accuracy over standard quantile bootstrapping, I think by adjusting for the underlying distribution's skew and bias. I am looking ...
0
votes
0answers
28 views

Confidence interval on mean of mixture distribution with normal components

I have 50 sacks, each containing material from one of 10 varieties of hemp. Each sack has a different weight. I don't know what variety is in each sack, or how much material I have of each variety. ...