By employing tweet data with location information for roughly one year in Japan, we observed the tweet base distribution of Twitter users. We compared the tweet base distributions to the country's population distribution based on the national census and identified a strong correlation. We also observed the daily time changes of the tweet-from-base probability and compared them to the daily changes of the awake-at-home probability that was investigated by the NHK Broadcasting Culture Research Institute. The patterns of both daily time changes are very similar, especially on weekdays.

The excitement and convergence of tweets on specific topics are well studied. We can also investigate the position-sensitive subjects by utilizing the position information of tweets. In this research, we focus on bomb terrorist attacks and propose a method for separately analyzing the number of tweets at the place where the incident occurred, nearby, and far. We made measurements of position-sensitive tweets and suggested a theory to explain it. This theory is the theory of social physics that can analyze epidemic phenomena such as movies and hit products from the time change of the number of social media writing. This paper expanded this theory to take into account the position where Tweet was transmitted. With this theory we found that it is possible to explain the spread of information by the tweets with location information on the bomb case.

We analyzed the historical data set of the Tokyo Stock Exchange (TSE) for a 17-month period from August 2014 to December 2015 which includes every transaction and order book snapshot, and found two major relationships: (i) a proportional relationship between the return of the market price and the order imbalance of the executed volume, and (ii) an inverse proportional relationship between the market impact and the averaged order book volume. In this analysis, we focus on daily and monthly time scale. We also introduce estimation results of transaction costs for a practical application of our studies.

By summing up the Point of Sale (POS) data of approximately 2,000 American supermarket stores from 2001 to 2012 for every company, we compared the growth rate of the POS sales data with each company's actual sales. We confirmed that the growth rate in quarterly sales for companies whose anchor products are daily necessities in the United States were strongly related to the POS data's growth rate, thus demonstrating that nowcast (real-time observation of company sales) is possible, at least for this type of business enterprise.

In this study, we investigated firm activity data in 2008 and 2014 in the United States and Japan. We reconfirmed that the decay rate of firm activity does not depend on firm age in Japan. This is consistent with the observation that firm age distribution follows an exponential function in Japan. At the same time, in the United States, we found that the decay rate of young firms is high and it becomes lower and settle a constant value as firms age. From these observations, we proposed the model that follows the property observed in the decay rate of firm activity. The constant decay rate in Japan is included as a particular case in this model. The model analytically leads to firm age distribution, the young firm in which deviates from an exponential function. This analytical feature is observed in empirical data in the United States. We confirmed the consistency of this analysis comparing parameters estimated from the decay rate with those from firm age distribution.