1: \begin{abstract}

2: 

3: In this paper, we prove that under proper conditions, bootstrap can further debias the debiased Lasso estimator for statistical inference of low-dimensional parameters in high-dimensional linear regression.

4: %consider the limiting distribution of the debiased Lasso estimator under conditions for sparsity pattern recovery with deterministic or Gaussian design. 

5: %Under such conditions, we prove that the limiting distribution of the debiased Lasso estimator has a substantial bias term, which can lead to a poor rate of convergence. Nevertheless, we consider applying Gaussian bootstrap to further ``debias'' the debiased Lasso estimator. We show that Gaussian bootstrap approximation is consistent under some regularity conditions. 

6: We prove that the required sample size for inference with bootstrapped debiased Lasso, which involves the number of small coefficients, can be of smaller order than the existing ones for the debiased Lasso. Therefore, our results reveal the benefits of having strong signals. Our theory is supported by results of simulation experiments, which compare coverage probabilities and lengths of confidence intervals with and without bootstrap, with and without debiasing.

7: \end{abstract}

8: