This paper presents a statistical characterization of a wideband directional indoor propagation channel based on measurements conducted in two indoor environments. Results show that multipath components (MPCs) arrive in clusters in both the spatial and temporal domains. The aim here is to investigate in more detail the clustering and correlation between these two domains. MPC parameters are estimated using the super-resolution frequency domain Space-Alternating Generalized Expectation (FD-SAGE) algorithm and clusters are identified in the spatio-temporal domain by a non-parametric density estimation procedure. The description of the clustering observed within the channel relies on two classes of parameters, namely, inter- and intra-cluster parameters which characterize the cluster and MPC, respectively. All parameters are described statistically by a set of empirical probability density functions (PDFs) extracted from the measured data. The correlation properties are incorporated in two joint PDFs for cluster and MPC position, respectively. Channel power density spectra (PDS) are also derived and are shown to exhibit exponential and Laplacian functions in the delay and angular domains, respectively.