Questions tagged [stratification]

A sampling technique in which the population of interest is partitioned into subsets ("strata") based on characteristics known at all units before sampling.

Filter by
Sorted by
Tagged with
0
votes
0answers
16 views

what is the best sampling procedure to use? (stratified)

all, i am a little confused on what sampling method to use. i am doing a classification problem where i am trying to classify a person as having cancer or not. there is apparently huge variance in ...
0
votes
1answer
341 views

How to partition data with multiple categorical features?

I have a data set with size of 30k+, and it has several categorical features and also several numerical features. When I try to split the data into training/testing data set, I need to answer a ...
4
votes
1answer
79 views

Main options on how to deal with imbalanced data

As far as I can tell, broadly speaking, there are three ways of dealing with binary imbalanced datasets: Option 1: Create k-fold Cross-Validation samples randomly (or even better create k-fold ...
69
votes
5answers
72k views

Understanding stratified cross-validation

I read in Wikipedia: In stratified k-fold cross-validation, the folds are selected so that the mean response value is approximately equal in all the folds. In the case of a dichotomous ...
0
votes
0answers
14 views

Should I maximize the number of respondent (total or per strata), or the similarity between the sample and the general population?

I want to conduct a market research survey. I want to interpret results for both the general population, which is very large, and also for age and household income subgroups. With the limited budget ...
0
votes
1answer
402 views

Stratified Sampling on Random Forest

I am working on a project to detect crops from satellite images by prediction. To do so, I use Random Forest model. I discussed with some people about whether to give sample weights on each tree in ...
26
votes
1answer
26k views

Benefits of stratified vs random sampling for generating training data in classification

I would like to know if there are any/some advantages of using stratified sampling instead of random sampling, when splitting the original dataset into training and testing set for classification. ...
4
votes
2answers
11k views

When do you stratify an analysis versus including an interaction term?

I’m not very familiar with when and why you would stratify on a variable or set of variables in a regression analysis generally and would like to know what the issues are particularly in contrast to ...
0
votes
1answer
22 views

How to explain that leave-one-out cross-validation doesn't need stratification in the context of classification?

Stratification is usually defined as training data and testing data having the same distribution of class values. In leave-one-out cross-validation, each fold only has 1 instance. I know that it's ...
0
votes
0answers
12 views

Distribution of the grand mean (mean of means), and case of stratified sampling

I am having some trouble while estimating the distribution of the mean of means, according to the type of sampling. Basically, I sampled N=1000 sets of n=100 random numbers distributed as an uniform ...
1
vote
0answers
21 views

Randomized blocking “after the fact”

I have 5 treatments that I'm going to randomly assign to visitors to my website. I'll have 10,000 or more visitors. Is it statistically valid to analyze the results based on visitor characteristics ...
0
votes
0answers
19 views

Post Stratification & Calibration? - Adding weight's - CALMAR

Let's suppose that I want to compare the average spending of both groups and I have two groups with 2 features in each one: Each client is unique and CityCode has only 2 groups CA and CB and AgeCode ...
2
votes
0answers
77 views

Asymptotic variance of Metropolis-Hastings estimates on a disjoint subdivision of the state space

I'm running the Metropolis-Hastings algorithm on a state space $(E,\mathcal E)$ which can be disjointly subdivided into regions $E_1,\ldots,E_k$, $k\in\mathbb N$ ($k\approx1e5$). On $E$, I have a ...
0
votes
0answers
11 views

Stratified Sampling vs. Hand picking a “random sample”, and using Chi-Squared as justification. Dichotomous outcome methods question

Suppose I would like to evaluate "the proportion of potato farms in Idaho that use pesticides". (dichotomous yes/no outcome variable). Now let's assume I know the exact number of potato farms in Idaho,...
1
vote
1answer
39 views

Sample stratification by dependent variable in linear regression analysis

I have a theoretical question that I would like some guidance on. Is it ok to stratify a sample population by the dependent variable? Does this bias regression results? For example, I'm doing an ...
0
votes
1answer
71 views

Choosing strata in stratified log-rank test

Let's say I would like to see which factors affect the survival of patients suffering from cancer. And let's assume I have two variables: chemo ($0$ - if a patient is not treated with chemotherapy or ...
0
votes
1answer
326 views

Cluster Analysis on a Stratified Random Sample

I have a set $A$ that consists of 5 million objects. Each object has a color, size and cost. I'd like to do a cluster analysis, e.g. K-means, on cost. However, it is computationally not feasible to ...
2
votes
2answers
259 views

Why class stratified sampling is not compatible for naïve bayesian modeling, if sampling is used?

I read a book recently and it mentioned related to prior probability of naïve bayesian: "Since the probability of an outcome is calculated from the data set, it is important that the data set used ...
0
votes
0answers
6 views

test of difference (equivalence) between pairs of hazard ratios derived from two models

I have two stratified cox proportional hazards regression models each restricted to data on men and women, respectively. Both models have the same covariates and the data on men and women come from ...
4
votes
1answer
692 views

How to do a stratified nonparametric test?

I'm trying to use the "coin" (conditional inference) package to perform a stratified nonparametric test for difference in distribution (for count data). I tried a stratified Mann-Whitney-Wilcoxon ...
1
vote
2answers
1k views

Possible catch 22 with Neyman's optimal allocation

I am using stratified sampling and neyman's optimal allocation to compute the best sample size for each stratum. neyman's optimal allocation is given by the formula, $$n_h = n \frac{N_h * S_h}{\sum_i ...
3
votes
2answers
226 views

Stratified Sampling Allocation

You believe that the mean electricity usage is about twice as much for houses as for apartments or condominiums, and that the standard deviation is proportional to the mean so that S1 = 2S2 = 2S3. How ...
0
votes
1answer
150 views

Sample stratified according to region - how representative is this sample on the regional level?

I am doing a survey analysis of a specific country, and I am interested in comparing different regions of that country. Generally speaking (if possible), when a sample is stratified according to ...
0
votes
0answers
16 views

Experiment design: what's the difference between diff in diff and post-stratification?

what's the difference between diff in diff and post-stratification? When should we use diff in diff and when should we use post_stratification quasi experiment design?
2
votes
1answer
28 views

Analysing stratified random sample

Let's say I am analysing an experimental design with treatment and control group which was based on random sampling design stratified on a number of covariates. When analysing the data I see two ...
1
vote
0answers
68 views

Percents of Random Samples

I have three sets $A,B,C$ of sizes $N_A=2508$, $N_B=36211$ and $N_C=2296$ respectively, containing binary values. I took 200 samples of each set to produce point estimates of the averages: $\hat p_A=0....
1
vote
0answers
34 views

Stratified sampling to generate random numbers (eg. for Monte-Carlo applications)

I am using a Monte-Carlo method to compute a value of interest $y$ from some input parameters $x_{i}$, that I use to draw statistical sets from simple distribution laws. In my case, for a single Monte-...
0
votes
0answers
15 views

Confidence interval on ratio of stratified sample estimates

I have a stratified sample I have conducted in all fifty states to determine the number of people with a characteristic of interest. (Let's say, purple people.) For illustrative purposes, in New ...
7
votes
2answers
3k views

Using post-stratification weights in R survey package

I am analyzing a dataset that has a variable for post-stratification weights. As this is a complex survey, the plan is to use the R survey package. I have been ...
18
votes
4answers
2k views

Is leave-one-out cross validation (LOOCV) known to systematically overestimate error?

Let's assume that we want to build a regression model that needs to predict the temperature in a build. We start from a very simple model in which we assume that the temperature only depends on ...
1
vote
1answer
56 views

On improving the sensitivity of controlled experiments by utilizing pre-experiment data (CUPED)

I’m currently reading “Improving the sensitivity of online controlled experiments by utilizing pre-experiment data” by Deng et al. and struggling to derive a few equations from the paper. I would ...
0
votes
0answers
24 views

Should I use simple random sampling instead of stratified sampling when some strata have low counts?

I am designing a survey for which I would like to stratify "events" by state / province, creating $h=86$ strata for the particular dataset I'm working with. However, there are some strata with low ...
1
vote
1answer
1k views

Neyman allocation and proportional allocation

Suppose that a city has 90,000 dwelling units, of which 35.000 are houses, 45,000 are apartments, and 10,000 are condominiums. We want to estimate the overall proportion (p) of households in which ...
0
votes
0answers
83 views

Should I use stratify parameter for scikit-learn train_test_split while dealing with highly unbalanced dataset?

I have a dataset with over 200000 records. Only 400 of are positive, which makes the data highly unbalanced. I cannot collect more data. At first I trained a decision tree. I used StratifiedKFold and ...
0
votes
0answers
21 views

Variance of estimator of stratified sampling

Consider a stratified design composed of $H$ strata of size $N_h$, $h = 1, . . . ,H$. We want to estimate the population mean $μ_y$ of the characteristic $y$. Let $μ_{x,h}$ $h = 1, . . . ,H$ be ...
0
votes
1answer
31 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 (...
0
votes
0answers
25 views

Stratification of sample data is lowering my accuracy?

So I've got this trainingset, it has a bunch of stuff yada yada.. Main point is that there are two target variables that only occur once in the dataset. This means I can't stratify when sampling, I ...
1
vote
0answers
27 views

Simulate confidence intervals for the absolute difference between two hazard ratios

I am examining the time to an (recurrent) event in response to different medications between men and women. I am adopting a 'within-individual' design whereby each person's time spent on- and off- the ...
0
votes
0answers
50 views

Is doing oversampling on train set and undersampling on test set correct?

I have an imbalanced dataset (95% in class 0 and 5% in class 1) and I am using machine learning for classification. The AUC(Area under ROC curve) was high (about 0.86) but AUPRC(Area under precision-...
0
votes
1answer
336 views

Optimal multivariate binning where the cut-points must be the same for all observations

I have a large data set with many discrete and continuous variables. All the variables are present in every observation. I want to explain (the log of) one continuous variable using all the other ...
0
votes
0answers
19 views

Interaction results in two datasets are concordant, but stratified analyses are not

Let's suppose I want to explore the relationship between a variable (SNP, {0,1,2}) and a disease outcome (D, continuous value such as blood pressure), knowing that this is different in two groups, let'...
33
votes
2answers
18k views

Why use stratified cross validation? Why does this not damage variance related benefit?

I've been told that is beneficial to use stratified cross validation especially when response classes are unbalanced. If one purpose of cross-validation is to help account for the randomness of our ...
1
vote
0answers
108 views

multilevel regression and poststratification mrp

I have some survey data. I want to calculate a regression with a stratified sample of this data. Because of the stratification, the sample size gets really small. Due to this, the standard error of ...
0
votes
0answers
12 views

Estimating a population proportion and credible interval from a biased stratified sample

I am attempting to calculate a point estimate and the upper 95th percentile (one sided credible interval) for a defect rate (i.e. a population proportion) using the results of a biased stratified ...
2
votes
1answer
86 views

Determine sample size using reversed goodness of fit

I am asked to work on a specific problem in which I have to calculate certain expenditures for an industry, consisting of a population of about 400 companies. Although I already suggested to conduct a ...
0
votes
2answers
331 views

k-fold CV-scheme stratifying response and considering groups

I have the following small dataset (n~140): 1-8 samples per patient A small fraction of samples belonging to the negative class (no tumor), ~ 10-20% A larger fraction of samples belonging to the ...
0
votes
0answers
8 views

How fine is the finest cluster level in a sample with individual data?

When using software to estimate the variance or standard error of statistics calculated from complex survey samples with clustering and stratification, if the sample contains data on individuals, is ...
5
votes
1answer
2k views

Bootstrapping dataset with imbalanced classes

I am trying to build an ensemble model to classify dataset with imbalanced data, where some of classes have just a few samples. And, because of this dataset property, when I am doing re-sampling with ...
2
votes
1answer
235 views

Interaction variables work differently when population is split

Let's say I have two predictors to predict financial risk: Gender and shopping habits. Gender has levels of "Male" and "Female", while shopping habits has "quick shopper" and "slow shopper". I am ...