mbb_resample_data: Moving Block Bootstrap (MBB) Resampler

In RCtest: Reality Check and Predictive Ability Tests for Forecast Evaluation

Moving Block Bootstrap (MBB) Resampler

Description

Generates a bootstrap resample of a time series matrix using the Moving Block Bootstrap (MBB) method of Kunsch (1989).

Usage

mbb_resample_data(data_series, block_length)

Arguments

data_series	matrix where rows are observations (P) and columns are variables (K).
block_length	integer block length for the resampler. Overlapping blocks of this length are sampled with replacement. A commonly used rule of thumb is block_length \approx T^{1/3} (Politis & Romano, 1994). For P = 165, this gives approximately 5–6.

Details

Resamples overlapping blocks of block_length consecutive rows with replacement, then concatenates them to produce a bootstrap sample of the same length P as the original series. This preserves the short-run autocorrelation structure of the data, which is required for valid inference in the Reality Check and SPA-type tests.

Value

matrix of bootstrap resampled data with the same dimensions as data_series.

References

Kunsch, H. R. (1989). The jackknife and the bootstrap for general stationary observations. The Annals of Statistics, 17(3), 1217–1241. \Sexpr[results=rd]{tools:::Rd_expr_doi("10.1214/aos/1176347265")}

Politis, D. N., & Romano, J. P. (1994). The stationary bootstrap. Journal of the American Statistical Association, 89(428), 1303–1313. \Sexpr[results=rd]{tools:::Rd_expr_doi("10.1080/01621459.1994.10476870")}

Corradi, V., & Swanson, N. R. (2011). The White Reality Check and some of its recent extensions. In Festschrift in honor of Halbert L. White.

Liu, R. Y., & Singh, K. (1992). Moving blocks jackknife and bootstrap capture weak dependence. In R. LePage & L. Billard (Eds.), Exploring the Limits of Bootstrap (pp. 225–248). Wiley.

Examples

data(metals)
# metals: 165 x 15; columns 1-3 are the first three competing forecasts
mbb_resample_data(metals[, 1:3], block_length = 5)

RCtest documentation built on June 2, 2026, 9:07 a.m.