1: \begin{abstract}

2: The \textit{adapted Wasserstein distance} ($\AWD$-distance) controls the calibration errors of optimal values in various stochastic optimization problems, pricing and hedging problems, optimal stopping problems, etc. However, statistical aspects of the $\AWD$-distance are bottlenecked by the failure of \textit{empirical measures} (\emp) to converge under this distance. Kernel smoothing and adapted projection have been introduced to construct converging substitutes of empirical measures, known respectively as \textit{smoothed empirical measures} ($\mathsf{S}$-\textit{Emp}) and \textit{adapted empirical measures} ($\mathsf{A}$-\textit{Emp}). However, both approaches have limitations. Specifically, $\mathsf{S}$-\textit{Emp} lack comprehensive convergence results, whereas $\mathsf{A}$-\textit{Emp} in practical applications lead to fewer distinct samples compared to standard empirical measures.

3: 

4: In this work, we address both of the aforementioned issues. First, we develop comprehensive convergence results of $\mathsf{S}$-\textit{Emp}. We then introduce a smoothed version for $\mathsf{A}$-\textit{Emp}, which provide as many

5: distinct samples as desired. We refer them as $\mathsf{AS}$-\textit{Emp} and establish their convergence in mean, deviation and almost sure convergence. The convergence estimation incorporates two results: the empirical analysis of the \textit{smoothed adapted Wasserstein distance} ($\AWDsigma$-distance) and its bandwidth effects. Both results are novel and their proof techniques could be of independent interest.

6: 

7: 

8: \noindent\emph{Keywords:} adapted Wasserstein distance, empirical measure, convergence rate, kernel smoothing\\

9: MSC (2020): 60B10, 62G30, 49Q22

10: \end{abstract}

11: