Feite Zhou

Financial cognitive ability is an important part of human capital, which has a direct impact on household income. The dual differences composed by rural residents’ cognition level and residents’ income make the relationship between the two complicated. Based on the analysis of the impact of financial cognitive level on rural household income, this paper takes into account the individual heterogeneity of financial cognition to build an unconditional quantile regression (UQR) model through the micro data of the China Household Financial Survey Project (CHFS). Incorporating the background of the rapid development of digital finance, this paper also studies the impact of rural residents’ financial cognition on household income. The empirical results of the article have the following conclusions: First, on the whole, the improvement of financial cognitive level promotes the increase of rural household income, but this promotion has significant differences at different quantile levels of household income. Second, the digital financial inclusion index in the region has a positive moderating effect on the impact of financial cognition on the income of rural households. Third, the heterogeneity of the impact of financial cognition on the income of rural households is reflected in the two aspects of regions and household assets.

The ups and downs of stock indexes are one of the most concerned issues for investors in the stock market. To improve the accuracy of stock index prediction, this paper compares traditional K-Fold Cross-Validation (KCV) and three cross-validation methods in the Radial Basis Function Support Vector Regression (RBF-SVR) model. They are named as Abandon Tail Cross-Validation (ATCV), Sequential Division Cross-Validation (SCV) and Gap Sequential Division Cross-Validation (GSCV). It is found that KCV has very limited validation ability for stock indexes time series data with no certain relevance. However, SCV and GSCV with small gap perform better, with high accuracy about 88% and small error about 2%. This research shows that the establishment of time series forecasting models for stock indexes needs to pay more attention to cross-validation methods, which cannot randomly dividing training set and test set. It is strongly recommended to use SCV and GSCV instead of KCV. In addition, the choice of the penalty parameter C and the radial basis kernel function parameter γ largely determines the accuracy and reliability of RBF-SVR stock index prediction model.