Back to search

Institutional Repository of Vilnius University

Stock price evaluation of streaming services through regression analysis of user growth and expansion models /

Abstract

dc:description

STOCK PRICE EVALUATION OF STREAMING SERVICES THROUGH REGRESSION ANALYSIS OF USER GROWTH AND EXPANSION MODELS Prof. Petras Dubinskas Bachelor thesis prepared - 2025 Vilnius The thesis consists of 54 pages. Number of tables at work - 46 pcs. Number of pictures at work - 0 pcs. Number of literature and sources - 68 pcs. Short description: This bachelor thesis analyzes stock price evaluation of three streaming companies – Netflix, Amazon and Disney – using descriptive analysis of expansion models and regression analysis of user growth. The work links subscribers, key business segments (AWS, Disney+) and past price movements with long-run stock prices. Problem, goal, and tasks of the work: The main problem is how to evaluate stock prices of streaming businesses whose value depends on both user growth and diversification into new activities. The goal is to identify which indicators best explain stock price movements of Netflix, Amazon and Disney. The tasks are: to review literature on international expansion and valuation; to describe the development of the three companies’ streaming services; to collect annual data on users, segment revenues and stock prices; to build simple linear regression, ARMA and moving average models; to test their assumptions and accuracy; and to compare results across companies. Research methods used in the work: Qualitative analysis; quantitative analysis: simple linear regression, ARMA models with 1 and 2 lags, moving average models with 2-, 3- and 4-years smoothing; diagnostic tests for homoscedasticity, autocorrelation, normality of residuals and multicollinearity. Calculations were performed in Excel – Data Analysis Toolpak and manual formulas. Research and results obtained: For Netflix, the main regression model with a dummy variable explained 91.43% of stock price variation; the best ARMA model with 1 lag achieved an even higher R^2 of 93.88%. For Amazon, the most reliable regression used AWS revenue and reached R^2=81.54%; the ARMA model with 1 lag and dummy variable performed best with R^2=92.50%. For Disney, the initial regression was weak, but after adding a dummy variable the model based on Disney+ users explained 91.40% of price variation, while ARMA models remained statistically insignificant. Moving average models showed that a 4-years window is the most accurate for all three companies: the lowest MAPE values were 31.00% for Netflix, 20.00% for Amazon and 18.00% for Disney. Conclusions: The results show that subscriber growth is a key driver of stock prices for Netflix and Disney, where adjusted regression models explain more than 91% of variation. For Amazon, AWS revenue is a stronger and more stable explanatory variable than the number of Prime users. ARMA models with 1 lag generally outperform simple regressions in terms of R^2, which means stock prices are also influenced by their own past level. Moving average models with 4-years windows give the best purely technical forecasts, but they smooth recent movements and should be used together with fundamental indicators. Overall, stock prices of streaming companies depend on a combination of user growth, performance of strategic segments and historical price movements, and the proposed models can be used to generate practical price forecasts for 2026.

Degree

thesis:*
Grantor dc:publisher
Institutional Repository of Vilnius University
Year dc:date
2026

Author and committee

dc:creator, dc:contributor.*
Author dc:creator
  • Chukhnii, Rostyslav,

Subjects

dc:subject × 1

Rights

dc:rights
Statement dc:rights
  • info:eu-repo/semantics/openAccess
Language dc:language
eng

Identifiers

dc:identifier.*

Chain of custody

source
Harvested from
Vilnius University
Base URL
epublications.vu.lt/oai
Last updated
2026-11-19
Source record
OAI-PMH GetRecord
citation

Chukhnii, Rostyslav,. Stock price evaluation of streaming services through regression analysis of user growth and expansion models /. Institutional Repository of Vilnius University, 2026. https://repository.vu.lt/VU:ELABAETD277692406&prefLang=en_US