Research on Complex Distribution Fitting and Evolution Model of Dynamic Data
DOI:
https://doi.org/10.62051/ijgem.v10n1.06Keywords:
Dynamic data, Distribution fitting, Evolution model, Parameter estimation, Time series analysisAbstract
In the era of big data, dynamic data in the fields of financial transactions, environmental monitoring, medical care and health care are growing explosively. Due to its timeliness, volatility and multi-source characteristics, the data distribution presents a complex multi peak and non-stationary state, and the traditional static fitting method is difficult to meet the actual analysis needs. This paper focuses on the complex distribution fitting and evolution mode of dynamic data, analyzes the core characteristics of dynamic data and the difficulties of distribution fitting, sorts out the applicable scenarios and improvement strategies of common distribution fitting models, constructs the representation system and identification method of evolution mode, verifies the effectiveness of the model through the real scene demonstration, and expands the application scenarios to complete the effect verification. The research shows that the improved dynamic fitting method based on Gaussian mixture model has significant advantages in multimodal data fitting, and the goodness of fit is improved by 10%~20% on average compared with the traditional method, and the evolutionary pattern recognition can effectively capture the trend and mutation of data, and the conclusion is consistent with the empirical research conclusion in the same field. This paper provides method support for accurate analysis of dynamic data, and has important reference value for decision optimization in the actual field.
Downloads
References
[1] Diaz-Rozo J, Bielza C, Larrañaga P. Clustering of data streams with dynamic Gaussian mixture models: An IoT application in industrial processes [J]. IEEE Internet of Things Journal, 2018, 5(5): 3533-3547.
[2] Young P C. Nonstationary time series analysis and forecasting [J]. Progress in Environmental Science, 1999, 1: 3-48.
[3] Gepperth A, Pfülb B. Gradient-based training of gaussian mixture models for high-dimensional streaming data [J]. Neural Processing Letters, 2021, 53(6): 4331-4348.
[4] Bishop C M, Nasrabadi N M. Pattern recognition and machine learning [M]. New York: springer, 2006.
[5] Burnham K P, Anderson D R. Multimodel inference: understanding AIC and BIC in model selection [J]. Sociological methods & research, 2004, 33(2): 261-304.
[6] Introduction to time series and forecasting [M]. New York, NY: Springer New York, 2002.
[7] Shumway R H, Stoffer D S. Time series analysis and its applications: with R examples [M]. New York, NY: Springer New York, 2006.
[8] McFarlane C, Raheja G, Malings C, et al. Application of Gaussian mixture regression for the correction of low cost PM2. 5 monitoring data in Accra, Ghana [J]. ACS Earth and Space Chemistry, 2021, 5(9): 2268-2279.
[9] Wu Y, Ni J, Cheng W, et al. Dynamic gaussian mixture based deep generative model for robust forecasting on sparse multivariate time series [C]//Proceedings of the AAAI Conference on Artificial Intelligence. 2021, 35(1): 651-659.
[10] Li L, Sun J, Ruan L, et al. Time-series analysis of continuous glucose monitoring data to predict treatment efficacy in patients with T2DM [J]. The Journal of Clinical Endocrinology & Metabolism, 2021, 106(8): 2187-2197.
Downloads
Published
Issue
Section
License
Copyright (c) 2026 International Journal of Global Economics and Management
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
How to Cite
