This study employs a stochastic volatility model to analyze hourly data during 2024 for seven major cryptocurrencies. The model incorporates explanatory variables derived from trading volume and returns, and Bernstein polynomials are used to flexibly capture intraday and intramonth seasonalities. Estimation results show that the regression coefficients for these explanatory variables are not statistically significant. This suggests that the typical relationships among volatility, returns, and trading volume observed in traditional financial markets may not hold at the hourly frequency in cryptocurrency markets. Although the volatility process displays strong persistence, no significant negative association is found between returns and volatility, indicating the absence of a leverage effect. Moreover, there is no clear evidence of seasonalities or fat tails in the volatility distribution. These findings imply that cryptocurrency markets are heavily influenced by algorithmic trading, and that price dynamics may not fully reflect the risk-averse behavior often attributed to human investors.

We forecast the net sales of individual manufacturers using scanner data. Since scanner data record product sales at retail outlets, it is reasonable to assume that sales forecasts can be derived from such data for manufacturers whose business performance is predominantly driven by domestic retail sales. Firms for which scanner data account for only a minor share of total sales are excluded from the scope of analysis. A previous study reported that the growth rate in quarterly sales for manufacturers whose anchor products are daily necessities in the United States were strongly related to the scanner data’s growth rate. In this previous study, they used all scanner data, but various error factors exist in scanner data and sampling bias is also large. Therefore in our study, in addition to forecasting sales using all scanner data, we cleanse scanner data before forecasting sales and verify the effectiveness of cleansing. We also examine the applicability of scanner data to stock investment.

We summarize modelling and estimating two types of dynamic skew-t copulas. The first one is for the dynamic generalized hyperbolic skew-t copula proposed by Ito and Nakamura (2019). The second one is for the dynamic Azzalini—Capitanio skew-t copula proposed by Ito and Yoshiba (2025). After summarizing the methods, we conduct empirical analysis using the fifteen-years weekly stock returns for two groups composed by three sectors from TOPIX 33 sectors. In results, we show the significance of the tail dependence and asymmetry and dynamic modelling for correlation matrix is more effective than static correlation matrix in terms of information criteria.

Indices for performance evaluation have been developed in line with the expansion of the crypto asset market, but these indices are currently based mainly on market capitalization-weighted (CW) indices. In the equity asset class, previous empirical studies have pointed out that the minimum variance (MV) portfolio is superior to the CW in terms of risk-return efficiency, and new indices have been developed. Asset management based on minimum variance indices is now one of the most popular investment strategies. With regard to crypto assets, although there are many studies on the effectiveness of adding crypto assets to a portfolio of traditional assets, there are few studies on portfolio construction within the crypto asset class, and to the authors’ knowledge, there are no studies comparing MV and CW yet. Therefore, in this study, we perform a comparative risk-return analysis of MV and CW in the crypto asset market. Monthly return data of 130 assets from Binance were used to construct MV. Given the high dimensionality and limited sample size, we used an improved version of Bayesian graphical LASSO which Oya and Nakatsuma (2022) proposed for estimation. 3-year operational experiments were conducted from 2022 to 2024, and MV was compared to a representative CW, S&P Cryptocurrency Broad Digital Market (BDM) Index. The results show that MV achieves higher returns with lower risk than CW, and is superior in terms of risk-return efficiency. This result is consistent with previous studies showing the superiority of MVs in the stock market and indicates that a similar trend may exist in the crypto asset market.

YUIMA is the name of a project for software implementation of statistical analysis methods for data modeled by stochastic differential equations (SDEs). In this paper, we explain how to implement statistical modeling of financial time series by SDEs using the R package yuima, which is the main development result of the YUIMA project. As a concrete example, we focus on daily time series of the exchange rate between the U.S. dollar and the Japanese yen from 2001 to 2024. The model selection results suggest that the volatility-induced stationarity observed in previous studies on the modeling of the U.S. short-term interest rate might also exist in the USD/JPY exchange rate.