Questions tagged [subsampling]

Subsampling is a resampling procedure akin to the bootstrap in which fewer than all observations are being drawn with replacement (vs. the original sample size used in the textbook bootstrap method). For creating samples out of your existing data, please consider "sampling" tag instead.

Filter by
Sorted by
Tagged with
2
votes
0answers
9 views

Subsampling the “right” amout of data to train an ML model

I am training a machine learning model (i.e., a classifier) on a large dataset. I know that I can get the same results using less data (about 30%) but I would like to avoid the trial and error process ...
0
votes
0answers
10 views

Distribution of a rate in a subsample after permutation of the total dataset

I need to estimate the distribution of a rate in a subsample after permutation of the total dataset. I was wondering whether it's sufficient to use the binomial distribution with the global rate ...
0
votes
0answers
25 views

Independent random sampling without replacement for sampled Gaussian mechanism

I was reading a paper about "Sampled Gaussian Mechanism" which is used in training differentially private ML models. This mechanism is a composition of ...
0
votes
0answers
20 views

comparing the effect and magnitude of two coefficients across two models

I have a panel dataset with 500 banks (in 20 countries) over the period 2000-2015. First, I perform a baseline regression by regressing my dependent variable; bank risk with my independent variable; ...
0
votes
0answers
15 views

R: repeated subsampling for unbalanced design with zero-inflated count data using glmmtmb

I have count data with the number of individuals per observation of 64 different plant species. 3 - 5 observations were made of each plant species, and the number of arthropods on the plants were ...
0
votes
0answers
26 views

How to treat apples and oranges best in multiple regression analysis? One fits all regression model or separate sub models?

I have a data set ready for multiple regression analysis that consist of apples and oranges. Let's say the depended variable is fruit size and there is a bunch of independent variables (categorical &...
1
vote
1answer
23 views

Combining Sub-Samples for Factor Analysis?

I am a newbie on the site and a relative newbie to some of the analysis I am trying, so my apologies in advance for any rookie mistakes or for asking what might be obvious to others! Can I run a ...
0
votes
0answers
15 views

Should coefficients estimated over subsample sum to coefficient estimated over full sample?

Suppose I have time series data and I estimate $$y=a+b_1x_{1}+...+b_nx_{n}+e~~~(1)$$ To be clear, let's focus on $\hat{b}_1$. suppose I have data ranging from $t=\{1,2,...,K,...,T\}$ suppose I want to ...
1
vote
2answers
53 views

Estimate linear regression coefficients and standard errors using sub-samples of dataset

I am working with a very large dataset (250 million records) and I want to use linear regression to estimate how some variables are related to the outcome variable. I have a lot of categorical ...
3
votes
0answers
80 views

Finding a sub-population from dataset matching another target dataset

Let's say one has a finite collection of i.i.d. samples from an unknown source distribution $S=\{x_{i} | i \in [1,n_{S}], x_{i} \sim p_{X_{S}}(x)\}$. Where each $x$ is multidimensional and has ...
0
votes
1answer
33 views

Speed Up KNN and Maintaining Accuracy for Anomaly Detection

This question is about using KNN in the context of anomaly detection. If the training dataset is large(10 M data points), KNN will be slow. Is subsampling(i.e. use a small subset of original training ...
0
votes
1answer
49 views

Time series forecast where each measurment is already averaged and has a spread

I would like to forecast a time series consisting of time averaged (everything happening during 15min intervals is averaged and recorded with a timestamp of the start of messurment) quantities (...
0
votes
0answers
17 views

Subsampling for odds ratio estimation?

I want to know how valid is to use a subsampling bootstrap for estimating the odds ratio. Here are the details: I have a large sample ($N \approx 7\times10^6$ ) with cases and controls ($cases/...
5
votes
1answer
170 views

Intuition behind m-out-of-n bootstrap

I am trying to get some intuition on why m-out-of-n bootstrap works but haven't been able to find good explanation. I would really appreciate any input on this. I think I do understand what bootstrap ...
0
votes
0answers
62 views

Can you use the isolation forest algorithm on large sample sizes?

The original isolation forest paper states that the algorithm works best on small subsamples, but is it okay to use it on large sample sizes or are other anomaly detection algorithms better?
0
votes
1answer
32 views

How to partition a sample into representative subsamples?

The problem is the following: take a sample $X$ of the general population $\Omega$. Each element of $\Omega$ (and hence, each element of $X$) is described by a vector of characteristics, each ...
0
votes
0answers
11 views

Determining how many raw data items to use in machine learning training

Firstly, I'd like to mention that I'm not a statistician or a machine learning expert. I am hoping to find a starter place or advice from ML experts/statisticians here to solve a problem related to ...
1
vote
0answers
34 views

Fitting glmms to data with very unbalanced sampling effort - could subsampling help?

I have a data set comprising measurements of invertebrate species richness from grab samples of seabed sediment collected from a shallow coastal area over a period of 20 years. The sampling effort is ...
0
votes
1answer
34 views

Subsampling large dataset before testing?

I'm running a Kruskal test with Conover post-hoc test to assess if there is a statistically significant difference between a numerical and a categorical variable with R. I previously created boxplots ...
1
vote
1answer
47 views

Regression - Interpretation of coefficients and probability

I am very confused about the output of my regressions. First of all, I am not even sure if I could divide my sample as I did, meaning that by subsampling as I did the variable ESG score is both ...
0
votes
0answers
20 views

Techniques to account for differences in misclassification “cost” on variables other than the outcome

Suppose you're in a classic classification context: you want to predict whether a patient has a certain virus. You are working in multiple regions (let's say 2 for simplicity: Region A and Region B) ...
1
vote
0answers
78 views

Covariance between sample mean and subsample mean

To estimate $\overline{Y}$(population mean), suppose an SRSWOR of size $n$ is taken from a population of size $N$, and the sample mean $\overline{y}$ is calculated. Then an SRSWOR subsample of size $...
0
votes
0answers
176 views

How to compute ESS (Effective Sample Size)?

I implemented the ESS calculation according to this manual like this: ...
1
vote
0answers
35 views

appropriate error term for a randomized complete block design, possible split plot in time, with subsamples (in SAS PROC MIXED))

I am analyzing the results of an agricultural experiment, as follows: Plots are laid out in a randomized complete block design with 3 replications. They have been either conventionally or organically ...
0
votes
1answer
42 views

How to perform inference on stratified sampling data

Let's say I'm studying a population of generic emergency calls to over the course of several months, and keeping track of the following independent variables: month (when the call happened) country (...
1
vote
1answer
135 views

Finding the variance of subsample-based estimation

Say, we know that the probability of an object having some property equals exactly $P$. We are given a sample (of size $N$) of these objects - in fact, that is a Binomial distribution with probability ...
0
votes
0answers
12 views

Presenting results of estimation with sub-sampling

I have a dataset that we can partition between three groups: Controls, Treated 1 and Treated 2. I want to run regressions that include the whole Control and Treated 1 groups, but I draw a random ...
2
votes
1answer
313 views

Do both Bootstrap with and without replacement create a distribution?

I'm having a "noisy debate" with colleagues about whether sampling without replacement can still create a distribution. Methodology: A bootstrap (iterative process where I calculate Somers' D for new ...
1
vote
0answers
24 views

Can I take a random sample of my very large data set to overcome non-independence?

I am trying to run a regression model on a very large time series data set (comparing flow noise to vehicle speed, pitch and dive state). Because my samples are taken about every minute (with some ...
1
vote
0answers
202 views

Subsampling as a method for time series train/validation splits

I have a question concerning train-test splits for time series data: Background I have a dataset of sensor data points for 13 month with datapoints measured every 5 minutes which I downsample to ...
0
votes
0answers
46 views

General way to construct a confidence interval for a unknown constant to which a sample estimator converges

Assuming that a sample estimator converges to some unknown constant (a wild assumption to be sure) and without assuming the distribution of either the sample estimator or the variables from which it ...
7
votes
4answers
1k views

Does sampling from a large dataset lead to correct inferences?

Say we have some population, and we obtain a "representative" random sample of that population, $(y_i, x_i)_{i = 1}^n$, where $n$ is very large (millions) and $x_i = (x_{i1}, x_{i2}, ... x_{ip})'$ is ...
0
votes
1answer
115 views

Statistic to Verify Subsample is Similar to Original Sample?

I have a subsample of data (120 students) that was taken from an original sample of 1,216 students' data. I need to report in my manuscript whether my subsample's key demographics (age, gender ...
0
votes
0answers
118 views

Subsampling to account for spatial autocorrelation of observations

I'm wondering to what extent (if any) subsampling of observations can be used to account for spatial autocorrelation within data. Is taking a smaller sample (subsample) of observations (without ...
0
votes
0answers
44 views

Run a regression with data from different measurements

I have a population $N$, that can be divided into several samples. 1 sample, $S$, was taken out from $N$, let's say a piece of paper from a page of a manuscript; this piece of paper is divided into ...
0
votes
1answer
1k views

What is the typical size of feature matrix for xgboost

In other words, I have a binary classification problem with million samples and around 1000 features. I am trying to understand wheather I should subsample the dataset and add a feature selection step ...
1
vote
1answer
2k views

How often to subsample for classification?

It is often recommended to subsample randomly if class sizes are unbalanced in classification - especially when classification accuracy is used. My question however: How often should the subsampling ...
2
votes
1answer
500 views

How to: Normal sub-sampling out of a uniformly distributed data samples

Given a uniformly distributed sample of data, It's needed to sub-sample out the points in a Normal distribution fashion, i.e. more around mean and sparser as we move out. What could be the steps?
0
votes
1answer
666 views

testing for differences using jackknife distributions

I have two distributions relative to two experimental conditions. I compute a certain index (i.e. coherence) describing each distribution. I want to see if there is a significant difference ...
5
votes
1answer
178 views

What is a good introductory text on resampling methods? [duplicate]

I have found a few decent ones about specific resampling applications such as bootstrapped confidence intervals, but nothing broader. A journal article or book chapter would be preferable to an entire ...
0
votes
0answers
46 views

Choosing subsample size (helping a friend analysing a smaller data set) [duplicate]

A friend of mine is working analysing 2000 twits per day and categorize them as postive, negative or neutral. This is a really boring task but the algorithms that do this classification are not very ...
11
votes
1answer
2k views

Chance that bootstrap sample is exactly the same as the original sample

Just want to check some reasoning. If my original sample is of size $n$ and I bootstrap it, then my thought process is as follows: $\frac{1}{n}$ is the chance of any observation drawn from the ...
1
vote
0answers
17 views

Sub-sampling to quantify mean and errors

I have a model in matlab that, ultimately, outputs a yes or a no depending on certain input parameters. There's a degree of randomness in the model, so by runnning it 1000 times, I may end up with 200 ...
1
vote
0answers
814 views

Subsampling - Choice of Subsample Size?

I have a question with regard to Subsampling. Subsampling: take samples without replacement of size b from the original sample of size n with b < n Bootstrapping: take samples with replacement of ...
6
votes
1answer
707 views

Is it good practice to perform model parameter tuning on a random subsampling of a large dataset?

A lot of the datasets presented to us in the company at which I'm currently an intern are very large (many millions of rows / Gigabytes, or even Terabytes of data). While running machine learning ...
1
vote
0answers
44 views

Significance for the whole dataset of detections in a subsample

I have a fairly simple question, but I am not so familiar with these problems so I would like to ask. If I have, let's say, 4 data samples belonging to the same overall population. If in a test a ...
0
votes
1answer
439 views

Repeated k-fold CV of sub sample - repeat the k-fold CV or repeat the sub sampling?

I want to do support vector regression using repeated k-fold cross-validation on a large dataset of 30k points. Because I do need to do a lot of those regressions I want to downsample the data first ...
5
votes
1answer
2k views

What is the effect of using survey sample weights for a sub-sample?

If a sub-sample of the survey sample, selected based on certain demographic characteristics of the data (e.g. age, race etc.), is used, which means the sub-sample might not be representative of the ...
0
votes
1answer
123 views

Stationarity of subsample

Consider that I have a weakly stationary series for the period 2003M1-2014M12. I want to make a VAR model for the subsample 2007M1-2014M12. Should I reconsider the weak stationarity of my series, so ...
2
votes
1answer
990 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)....