Loading...
| Friend's email | |
| Your name | |
| Your email | |
| enter code | |
This page was sent successfuly
- Type of Document: M.Sc. Thesis
- Language: Farsi
- Document No: 57056 (02)
- University: Sharif University of Technology
- Department: Mathematical Sciences
- Advisor(s): Haji Mir Sadeghi, Mir Omid
- Abstract:
- In the era of the internet and with the emergence of online stores, the possibility of changing prices for these e-commerce platforms has become simple and virtually cost-free. This allows them to adjust prices optimally in response to environmental and surrounding factors, maximizing their revenue from product sales. This development has led to numerous applications of statistical learning methods, particularly online learning, in these markets. In this thesis, we explore the issue of online pricing from various perspectives. One aspect is that product pricing is based on their features, which can be numerous. This leads us to the use of online learning methods and statistical learning in high dimensions. Another perspective is the learning of the demand function. In all these methods and statistical techniques, there is a seller who receives distinct products online and prices them with the goal of maximizing profit, without knowing the product's value to the customer or the optimal price for each product. However, the seller can learn this based on whether the product has been sold at previously announced prices or not. In this research, we use techniques from the field of statistical learning to understand and learn about the environment and the value of products for customers. We aim to find an algorithm that performs close to the optimal algorithm. In this study, our criterion for evaluating the efficiency of pricing algorithms is regret. We first introduce algorithms that assume the value of a product is linear based on its features and prove sub-linear regrets over time for them. Then, assuming a limitation on the number of price changes, we introduce another algorithm which also achieves sub-linear regret using statistical learning techniques. The more frequent the price changes, the lower this regret will be
- Keywords:
- Dynamic Pricing ; Statistical Learning Method ; Online Learning ; Value Function ; Optimal Price ; Stochastic Analysis ; Feature Vector ; Regret Criterion ; Linear Optimization
Digital Object List
محتواي کتاب
Bookmark