This paper first carries on the elaboration to the least squares support vector machine (LS-SVM) forecast model. Basing on the theory of wavelet frame and the condition of the SVM kernel function, a method that generates wavelet kernel function of the support vector machine is proposed. Then the Mexican Hat wavelet is been selected to construct LS-SVM kernel function and form LS-SVM model based on the wavelet kernel function (the LS-WSVM model), after that forecast freight traffic of highway by this model in China. Through the contrast of forecast result between four different kernel functions, it indicated that the model using wavelet kernel function have a higher validity than that of other kernel functions. At the same time, contrasting the results between LS-WSVM forecast and other forecast methods, it also indicated the LS-WSVM is able to increase the forecast precision. After taking the model into different areas of china, we find that the model has the higher application value
This paper proposes a kind of combination forecast model based on the wavelet analysis. First, the insurance income time series is decomposed under a multi-scale by using the Mallat algorithm. So the approximate (low frequency) component and the detail (high frequency) component under the corresponding scale are obtained. Then, the trend and cycle are distilled from the decomposition and modeled separately. After that the stochastic item is modeled. Finally, the superimposition of forecasts result can forecast the original insurance income. Through the contrast of forecast result with the tradition forecast model ARIMA by the data of Chinese insurance income, it indicates that this model makes full use of the existing information and it is more precise in forecast
Social media text offers new opportunities for tracking psychological constructs over time, but such measurement requires methods that are interpretable and stable in longitudinal settings. Grounded in Conceptual Space Theory and semantic projection, this study develops the Word Embedding Projection Approach (WEPA), which represents psychological constructs as semantic axes defined by theory-driven anchor words in a domain-specific word embedding space. WEPA scores are interpreted as indicators of expressed psychological salience in user-generated text. Using panel data from Keep, a Chinese fitness-oriented social media platform (177,829 users; 2,668,298 weekly observations), we evaluate WEPA across four studies. Study I examines the measurement foundation through anchor-word validation, human-coded benchmarks, and cross-temporal semantic-axis stability checks. Studies II and III show theoretically interpretable predictive associations between WEPA-derived scores and subsequent exercise duration for goal-setting constructs and the four source dimensions of self-efficacy. Study IV provides exploratory relative-week trajectories of aggregate construct dynamics. Overall, WEPA shows strong agreement with human-coded benchmarks for goal specificity and physiological states and achieves broader text coverage than a dictionary-based baseline. These findings suggest that WEPA offers a promising approach to theory-driven psychological measurement from domain-specific social media text.