Block Resampler

class tsbootstrap.block_resampler.BlockResampler(blocks: List[ndarray], X: ndarray, block_weights: Callable | ndarray | None = None, tapered_weights: Callable | ndarray | None = None, rng: Generator | Integral | None = None)

A class to perform block resampling.

resample_blocks()

Resamples blocks and their corresponding tapered_weights with replacement to create a new list of blocks and tapered_weights with total length equal to n.

resample_block_indices_and_data()

Generate block indices and corresponding data for the input data array X.

property X: ndarray

The input data array.

property block_weights: ndarray

An array of normalized block_weights.

property blocks: List[ndarray]

A list of numpy arrays where each array represents the indices of a block in the time series.

resample_block_indices_and_data()

Generate block indices and corresponding data for the input data array X.

Returns:

A tuple containing a list of block indices and a list of corresponding modified data blocks.

Return type:

Tuple[List[np.ndarray], List[np.ndarray]]

Example

>>> block_resampler = BlockResampler(blocks=blocks, X=data)
>>> block_indices, block_data = block_resampler.resample_block_indices_and_data()
>>> len(block_indices) == len(data)
True

Notes

The block indices are generated using the following steps: 1. Generate block weights using the block_weights argument. 2. Resample blocks with replacement to create a new list of blocks with total length equal to n. 3. Apply tapered_weights to the data within the blocks if provided.

resample_blocks()

Resample blocks and corresponding tapered weights with replacement to create a new list of blocks and tapered weights with total length equal to n.

Returns:

The newly generated list of blocks and their corresponding tapered_weights with total length equal to n.

Return type:

Tuple[list of ndarray, list of ndarray]

Example

>>> block_resampler = BlockResampler(blocks=blocks, X=data)
>>> new_blocks, new_tapered_weights = block_resampler.resample_blocks()
>>> len(new_blocks) == len(data)
True

property rng: Generator

Generator for reproducibility.

property tapered_weights

A list of normalized weights.