Talk · International Conference on Electronics, Engineering Physics and Earth Science · September 20, 2026
Comparative Study of Regression Models Using Machine Learning to Predict Real Stock Prices
Assel Mukasheva, Almas Saduakas, Dinargul Mukhammejanova, Anel Aidos, Күлжан Тогжанова
Abstract
Sale price prediction is an important task for breeders, sellers, buyers, and online equine marketplace platforms, as horse value is influenced by a combination of animal-specific and market-related characteristics. Unlike standardized agricultural commodities, horse prices vary substantially depending on breed, age, gender, height, color, location, riding discipline, and additional characteristics. Traditional horse valuation is often based on expert judgment or comparison with similar animals, which may lead to subjective and inconsistent price estimates. This study investigates the use of machine learning models for predicting horse sale prices based on online listing data collected over a ten-year period. After data cleaning, the final dataset contained 6010 horse sale records. Since the raw price distribution was strongly right-skewed, the base-10 logarithm of price was used as the regression target. Several supervised regression models were trained and compared, including Linear Regression, Ridge Regression, Random Forest, Gradient Boosting, XGBoost, and a Multi-Layer Perceptron. The results show that tree-based ensemble models outperform linear models, with Gradient Boosting achieving the best performance on the held-out test set. The findings indicate that horse price formation involves nonlinear relationships and interactions among features such as age, height, breed, discipline, location, and gender. This study contributes to the development of data-driven tools for more objective horse price estimation and decision support in equine markets.
More by Assel Mukasheva
- Performance Evaluation of WPA2 and WPA3 Security Protocols in Private Wi-Fi 6 Environments2026
- HYBRID GAT-AUTOENCODER-BILSTM-ATTENTION ARCHITECTURE FOR MULTI CLASS CYBERATTACK DETECTION IN HETEROGENEOUS IOT NETWORKSВестник Алматинского университета энергетики и связи · 2026
- DISTRIBUTED SYSTEM FOR DATA SHARING AND JOINT PROCESSING BETWEEN RESEARCH CENTERS AND FARMSВестник Алматинского университета энергетики и связи · 2026
- Research on Vulnerability Detection Methods in SQL and NoSQL Systems Using Machine Learning and Large Language ModelsInternational Conference on Electronics, Engineering Physics and Earth Science · 2026 · Citations: 1