We examine whether extension of trading hours can create incr eased trading activity and price efficiency by utilizing an agent agent-based market model. Specifically, we examine whether the partial extension of trading hours hours?that is, implementing the pre pre-market session and the after after-hours sessionsession?and what duration of the session is effective. The simulation result reveals that the implementation of both sessions could have a negative impact on trading activity and price efficiency if investors' participation during the session is limited; it could result in more concentrat ed trading in the opening session, wider divergence between market prices and the fundamental value, and lower price stability. In addition, longer sessions are less beneficial (or more harmful). However, we find that the implementation of the pre pre-market s ession is far more beneficial than that of the after after-hours session; specifically, the implementation of the short short-term pre pre-market session could induce higher price efficiency and higher price stability regardless of the number of market participants during the session.

The author integrated 1) multiple time-series analysis, 2) deep-learning, and 3) wavelet transform technics to forecast financial and economic time-series data. More specifically, the general purpose predictor was developed, which exploits large number of observable variables by summarizing them into latent factors through deep-learning. This can be regarded as a deep-learning version of Factor Augmented VAR model. As a preprocessing step, all observable variables are decomposed into cyclical components (waves) and a trend component by multiple resolution analysis based on "wavelet transforms". FAVAR model is fitted to the each decomposed series with extrapolating forecasts, and then integrated into the fitted values and forecasts of original series. The back-test for the period from Jan 2015 to Apr 2016 showed good performances of the 1-month- and 3-month-ahead predictions for TOPIX, USDJPY and other economic indicators, comparted with the simple VAR model.

In this paper, we introduce new approach to estimate firms' commercial networks based on their historical stock price. Recently commercial networks are keenly researched to expand the accuracy of risk management especially in estimating negative impacts from business counter parties. The standard approach to obtain networks is firstly we collect all firms' financial reports as of same timing with the corporation of all financial institutions, secondly extract business transactions by carefully checking details of their balance sheets, and thirdly unite each firms without any discrepancy. It would take enormous time and efforts. We utilize historical stock price to escape from having these difficulties. We introduce our approach with overviewing simple model guidance and explaining a few samples of numerical experiments.

Bond price is affected by main two factors; fundamentals,and forces of demand and supply. The latter impact on prices has a tendency to extinguish in less time compared to the former one. This papers provides an investment strategy focused on forces of demand and supply. We analyze a time series of bonds prices using Principal Component Regression and Partial Least Squares Regression to detect the forces of demand and supply that skews the equilibrium price and generates arbitrage oppotunities.

We propose a method of extracting basis information on analyst's forecasts from analyst reports. Our method extracts basis information on analyst's forecasts from analyst reports by using frequent expressions (e.g., "earning capacity") and clue expressions (e.g., "is expected"). The frequent expressions and clue expressions are extracted from the analyst reports automatically. We evaluated our method and it attained 75:0% precision and 61:7% recall, respectively.

In today's low interest rate environment, a central bank places importance on the communication with the markets about future monetary policy and it has been recently observed that the market prices are affected by forward guidance published by a central bank. Therefore, it is important to reveal the impact of the forward guidance on expectations of market participants and asset prices. In this study, we classified FOMC meeting minutes into 8 topics using LDA, and calculated sentiment value for each topic considering subtrees of dependency structure. As a result, we revealed that the sentiments extracted by our method were largely related to macroeconomic indicators comparing with conventional methods.

In this study, we propose the new economic standard which can estimate the value of financial assets without any use of spicified currency or asset. Time series of price is the most essential subject for financial studies and is supposed to show the changing value of the assets, but in reality, a price shows not only the value of the asset but also the value of currency. The volatility of currencies are none the smaller for that of assets, therefore, time series analysis contains inveitable distortion. In this paper we show some actual examples and how the new standard solves this problems.

Smart beta investments have recently enjoyed popularity among Japan market investors, leading to a substantial increase of Minimum Variance strategies in the form of ETF products. While each investment product, whether smart beta or traditional active fund, is designed to fulfill its investor expectations, many wonder whether there are any market effects from the trading of smart beta strategies on the more traditional strategies. I discuss if it is possible to find the background of phenomena and if there are methods to avoid the impact from smart beta strategies. I try to construct an optimal portfolio that seeks alpha by estimating a robust factor model and conduct a simulation that employs advanced techniques, including the use of residual variance scalars to correct for the Simultaneous Estimation Bias problem.