DATASET BOOTSTRAP SAMPLING¶
Performs bootstrap resampling on a dataset to estimate statistical properties (mean, standard deviation, or raw samples) for selected columns with a configurable confidence interval. Use this worker when you need robust distributional estimates or confidence bounds from an existing tabular dataset without making parametric assumptions.
When to use¶
Classification: process.
Tagged: bootstrap, confidence_interval, mean, monte_carlo, resampling, sampling, standard_deviation, statistics.
Inputs¶
| Label | ID | Type | Default | Required | Description |
|---|---|---|---|---|---|
| Dataset | dataset_1 | dataset | — | Input tabular dataset to be bootstrap-resampled; must contain at least the columns specified in the ‘columns’ input. | |
| Columns | columns | scalar | — | ✓ | One or more column names from dataset_1 on which bootstrap statistics will be computed; multi-select list populated dynamically from the connected dataset. |
| Number of Samples | num_samples | scalar | 5000 | Number of bootstrap resample iterations to draw; default is 5000 — increase for tighter confidence estimates, decrease to reduce runtime. | |
| Confidence Interval | conf_int | select | 90 | Confidence interval level (%) used to compute lower and upper bounds on the bootstrapped statistic; choose 90, 95, or 99 (default 90). | |
| Stat Type | stat_type | select | mean | Statistic to compute across bootstrap samples: ‘mean’ (default), ‘standard deviation’, or ‘samples’ to return the raw resampled values. | |
| Track Raw Samples | track_raw | select | no | Whether to retain all raw bootstrap sample values in the output (‘yes’) or return only the aggregated statistic and confidence bounds (‘no’, default). | |
| Return Type | return_type | select | mean |
Outputs¶
| Label | ID | Type | Description |
|---|---|---|---|
| Output | dataset_output | dataset | Output dataset containing the bootstrapped statistic (mean or std dev) and confidence-interval bounds for each selected column, or the full raw resampled values if track_raw is enabled. |
Disciplines¶
- data.dataset.transform
- data.statistics
- design_exploration.reliability
Runnable example¶
A runnable example is registered for this worker. Open the example workflow on the d3VIEW canvas: /api/workflow/example?id=dataset_bootstrap_sampling
Auto-generated from transformation schema. Worker id: dataset_bootstrap_sampling. Schema hash: 1a4be3a0bf7f. Hand-curated docs in workerexamples/ override this page when present.