Ryohei Hisano

Sales distribution of consumer electronics

Using the uniform most powerful unbiased test, we observed the sales distribution of consumer electronics in Japan on a daily basis and report that it follows both a lognormal distribution and a power-law distribution and depends on the state of the market. We show that these switches occur quite often. The underlying sales dynamics found between both periods nicely matched a multiplicative process. However, even though the multiplicative term in the process displays a size-dependent relationship when a steady lognormal distribution holds, it shows a size-independent relationship when the power-law distribution holds. This difference in the underlying dynamics is responsible for the difference in the two observed distributions.

The Gradual Evolution of Buyer-Seller Networks and Their Role in Aggregate Fluctuations

Buyer–seller relationships among firms can be regarded as a longitudinal network in which the connectivity pattern evolves as each firm receives productivity shocks. Based on a data set describing the evolution of buyer–seller links among 55,608 firms over a decade and structural equation modeling, we find some evidence that interfirm networks evolve reflecting a firm’s local decisions to mitigate adverse effects from neighbor firms through interfirm linkage, while enjoying positive effects from them. As a result, link renewal tends to have a positive impact on the growth rates of firms. We also investigate the role of networks in aggregate fluctuations.

A new approach to building the interindustry input-output table using block estimation techniques

We present a new approach to estimating the interdependence of industries in an economy by applying data science solutions. By exploiting interfirm buyer-seller network data, we show that the problem of estimating the interdependence of industries is similar to the problem of uncovering the latent block structure in network science literature. To estimate the underlying structure with greater accuracy, we propose an extension of the sparse block model that incorporates node textual information and an unbounded number of industries and interactions among them. The latter task is accomplished by extending the well-known Chinese restaurant process to two dimensions. Inference is based on collapsed Gibbs sampling, and the model is evaluated on both synthetic and real-world datasets. We show that the proposed model improves in predictive accuracy and successfully provides a satisfactory solution to the motivated problem. We also discuss issues that affect the future performance of this approach.

Semi-supervised Graph Embedding Approach to Dynamic Link Prediction

We propose a simple discrete-time semi-supervised graph embedding approach to link prediction in dynamic networks. The learned embedding reflects information from both the temporal and cross-sectional network structures, which is performed by defining the loss function as a weighted sum of the supervised loss from the past dynamics and the unsupervised loss of predicting the neighborhood context in the current network. Our model is also capable of learning different embeddings for both formation and dissolution dynamics. These key aspects contribute to the predictive performance of our model and we provide experiments with four real-world dynamic networks showing that our method is comparable to state of the art methods in link formation prediction and outperforms state-of-the-art baseline methods in link dissolution prediction.

The Gradual Evolution of the Interfirm Buyer-Seller Network and Its Role in Aggregate Fluctuations

The interfirm buyer-seller network is important from both macroeconomic and microeconomic perspectives. From a macroeconomic perspective, this network represents a form of interconnectedness that allows firm-level idiosyncratic shocks to be propagated to other firms. This propagation mechanism interferes with the averaging out process of shocks, having a possible impact on aggregate fluctuation. From a microeconomic perspective, the interfirm buyer-seller network is a result of a firms strategic link renewal processes. There has been substantial research that models strategic link formation processes, but the economy-wide consequences of such strategic behaviors are not clear. We address these two questions using a unique dataset for the Japanese interfirm buyer-seller network. We take a structural equation modeling, and show that a large proportion of fluctuation in the average log growth rate of firms can be explained by the network and that link renewal by firms decreases the standard deviation of the log growth rate.

A New Approach to Building the Interindustry Input--Output Table

We present a new approach to estimating the interdependence of industries in an economy by applying data science solutions. By exploiting interfirm buyer--seller network data, we show that the problem of estimating the interdependence of industries is similar to the problem of uncovering the latent block structure in network science literature. To estimate the underlying structure with greater accuracy, we propose an extension of the sparse block model that incorporates node textual information and an unbounded number of industries and interactions among them. The latter task is accomplished by extending the well-known Chinese restaurant process to two dimensions. Inference is based on collapsed Gibbs sampling, and the model is evaluated on both synthetic and real-world datasets. We show that the proposed model improves in predictive accuracy and successfully provides a satisfactory solution to the motivated problem. We also discuss issues that affect the future performance of this approach.The roles and interactions which nodes take part in has a significant impact on the structure of a network. In order to estimate this underlying latent structure, its numbers and composition from real data, flexible treatment of the structural uncertainty and efficient use of available information becomes a key issue. I take a Bayesian nonparametric approach, jointly modeling sparse network, node textual information and potentially unbounded number of components to handle the aforementioned task. I show using synthetic and real datasets that my model successfully learns the underlying structure utperforming previous method.