Essays in Price Measurement with Statistical and Machine Learning Methods
- Price indices play a vital role in economic measurement as they reflect price levels and measure price fluctuations. Price level measures are used with macroeconomic indicators to express them in real terms. These measures are also used to index wages, rents, and pensions. Furthermore, they are used as a reference for monetary policy conducted by central banks. Therefore, the provision of accurate price indices is one of the most important goals of National Statistical Institutes (NSIs), and numerous studies have been devoted to this goal. This cumulative dissertation also contributes to this goal. It contains four chapters, each of which represents a separate research. The first two studies are devoted to the treatment of seasonal products by using different price index methods. The first research is co-authored with Ken van Loon. The third research is dedicated to finding the most accurate method to make price predictions for missing products. The fourth research is focused on the treatment of products by using different price index methods when products’ quality characteristics are available.
| Author: | Botir Radjabov |
|---|---|
| URN: | urn:nbn:de:hbz:385-1-28013 |
| Referee: | Ludwig von Auer, Normann Lorenz |
| Advisor: | Ludwig von Auer |
| Document Type: | Doctoral Thesis |
| Language: | English |
| Date of completion: | 2025/12/22 |
| Publishing institution: | Universität Trier |
| Granting institution: | Universität Trier, Fachbereich 4 |
| Date of final exam: | 2025/12/11 |
| Release Date: | 2026/01/30 |
| Tag: | CPI; Seasonal Products; Multilateral Methods; Axiomatic Tests; Statistics; Machine Learning; GEKS and Weighted GEKS; Tree Based Methods; Regularization Methods; |
| Number of pages: | x, 122 Blätter |
| Institutes: | Fachbereich 4 |
| Licence (German): | CC BY-NC-SA: Creative-Commons-Lizenz 4.0 International |


