Questions tagged [sample-weighting]

The tag has no usage guidance.

Filter by
Sorted by
Tagged with
8 votes
1 answer
204 views

How does class balancing via reweighting affect logistic regression?

In machine learning approaches to classification, practitioners sometimes upweight the minority class to achieve a 50-50 balance, claiming that this results in improved classification performance. ...
  • 10.3k
0 votes
0 answers
22 views

How does a RandomForestClassifier in sklearn use sample weights?

How does a RandomForestClassifier in sklearn use sample weights? Are sample weights applied when Random Forest bootstraps? Are sample weights applied when model fitting i.e. when training the loss ...
0 votes
0 answers
17 views

Using entropy balancing to achieve general population representativeness in a survey with multiple conditions

We are currently planning a survey in which we ask people about their attitudes towards four different groups of people. Each participant is only asked about their attitudes towards one of the four ...
0 votes
0 answers
47 views

How to weight probabilites of getting sampled depending on the frequency of occurrence?

Unfortunately I'm struggling to describe my problem mathematically. I have 2000 strings, many of which are repeated. Now I want to write a 'random' sampling algorithm that produces 100 samples out of ...
  • 1
0 votes
0 answers
104 views

binary classification with labels of varying quality

I have a binary classification problem, where half my development data is labeled by a reliable source, and the other half by a less reliable one. Note that each instance gets a label from only one of ...
0 votes
0 answers
17 views

Weighting for modelling probability of selection

I want to use inverse probability weighting in some regressions and to estimate some weighted means from a non-representative sample. I plan to estimate a probit model for probability of selection ...
  • 113
1 vote
0 answers
13 views

Nonresponse weight adjustments in multi-stage household surveys

I have a question about nonresponse weighting in complex sample surveys in multi-stage designs, like say, The US National Comorbidity Survey Replication (NCS-R), the Health and Retirement Study (HRS), ...
3 votes
1 answer
649 views

Is IPTW (inverse probability of treatment weighting) legal?

When using IPTW, one can easily get weights 10 or even 20 for the observations. For instance, in logistic regression, weight 10 for an observation means that we have not one, but 10 observations ...
1 vote
0 answers
497 views

XGBoost regressor sample weight has negligible impact on performance

I am using XGBoost regressor for a prediction problem. I did a 70/30 split for the available data (around 60K samples) to split training/validation. For the training portion, I used 80% to train the ...
  • 111
0 votes
1 answer
110 views

Sample weighting vs. (e.g. one-hot encoded) categories

I have seen recommendations to use sample weighting when the training dataset is not evenly balanced over known categories, so that an imbalance in the number of elements in each category does not ...
1 vote
1 answer
104 views

WeightIT package error: treatment and covariates must have same number of units [closed]

While using the weightIt package in R I encountered a strange error: "error: treatment and covariates must have same number of units" Now, checking the root code of the package and this ...
2 votes
1 answer
64 views

Domain adaptation under covariate shift: estimating density ratio through a classifier

In domain adaptation under covariate shift, one approach is to weight the instances from the source domain by a factor $\frac{p_T(x)}{p_S(x)}$ in the training, where $p_S(x)$ and $p_T(x)$ represent ...
  • 866
1 vote
1 answer
19 views

How to handle different sized experiment samples

Imagine a 3 by 1 experiment. One group has 1,000 observations, one with 5,000 observations, and 4,000 observations in the last group. I'm trying to see whether the manipulation between groups in the ...
0 votes
1 answer
22 views

What is the sample size and variance for the mean that is a simple, unweighted average of two independent groups' means?

We have two groups: N1 = 10 and N2 = 100 Their means on some measurement are: Mean1 = 4 and Mean2 = 5 Their variances are Var1 = 3 and Var2 = 2.5. Let's further assume we have no access to the ...
  • 3
0 votes
1 answer
484 views

Including a weighting variable in a linear regression

I'm looking at how temperature affects length. My length variable is the mean length calculated for every year, it is derived from ~10,000 data points. Not every year had the same sampling effort (e.g....
2 votes
0 answers
35 views

How should one compute confidence intervals for means computed with inverse propensity weights (IPW)?

Inverse propensity weighing involves a machine learning model that takes features and outputs the predicted probability that this person is in the sample. Let $w_i$ be the inverse of the output for ...
  • 319
1 vote
1 answer
162 views

Weighted Survey Predicted Probabilities

I have a question about calculating prevalence using predicted probabilities from a survey weighted generalized linear model. Say my goal was to calculate the prevalence of a binary outcome using the ...
  • 122
3 votes
1 answer
729 views

Minimize SSE function

Consider a data set in which each target $t_n$ is associated with a weighting factor $r_n > 0$, so that the sum-of-squares error funtion becomes $$SE(w)= \frac{1}{2} \sum_{n=1}^N r_n \left(\...
3 votes
3 answers
238 views

Weighting significance tests according to the appropriateness of their assumptions

Consider a t-test of means. One formula for computing the p-value assumes equal variances. Another formula assumes unequal variances. With small sample sizes the tests can give quite different ...
  • 3,315