**6. Conclusions**

In this work, we have introduced an asymmetric definition for relatedness by extending the PCPG methodology introduced in [30] for its use on bipartite datasets, which we call biPCPG. We apply this approach to a recently introduced dataset containing the exports of countries regarding both manufactured products and intangible services. We show that the biPCPG methodology is able to generate a statistically robust network of economic sectors which captures the underlying influence structure int erms of temporal correlations.

This work can be extended in a number of possible directions. First of all, the biPCPG framework can be applied to any temporal bipartite network, such as those of common use in economic complexity, such as the company-technology [9] or the country-scientific field network [29]. Moreover, the adapted bootstrapping procedure can be used to other network-generating techniques based on correlation-filtering to datasets with a multisample and multi-variable structure. These techniques include those based on threshold methods [53], the Minimum Spanning Tree [33] and the aforementioned PMFG [31], as well as more recent techniques based on a null-model approach [54]. This would be possible by replacing the last step in our procedure, the original PCPG algorithm, with the correlation-filtering technique of interest. Finally, it would also be particularly interesting to apply our procedure to datasets with the same structure but longer time series, such as financial datasets containing, for example, asset prices at the different exchanges where they are traded.

**Author Contributions:** Conceptualisation, A.Z. and T.D.M.; methodology, A.Z. and T.D.M.; software, C.S.d.P.P.; validation, C.S.d.P.P., A.Z. and T.D.M.; formal analysis, C.S.d.P.P.; investigation, C.S.d.P.P., A.Z. and T.D.M.; data curation, A.Z.; manuscript writing, review and editing, C.S.d.P.P., A.Z. and T.D.M. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research received no external funding.

**Institutional Review Board Statement:** Not applicable.

**Informed Consent Statement:** Not applicable.

**Data Availability Statement:** Raw databases are available from (https://comtrade.un.org, accessed on 13 February 2019) and (https://data.imf.org, accessed on 13 February 2019).

**Acknowledgments:** The authors acknowledge Michele Tumminello for providing the PCPG Mathematica code.

**Conflicts of Interest:** The authors declare no conflict of interest.
