Webová aplikace na analýzu, predikce a rebalanci akciového portfolia v Pythonu

This master’s thesis focuses on the design and implementation of a web application for the analysis, forecasting, and rebalancing of an equity portfolio in Python. The theoretical part summarizes the basic principles of investment portfolios, performance and risk measurement, modern portfolio theory, asset allocation optimization methods, and financial time series modeling. Particular attention is paid to return and risk indicators, time-weighted return, volatility, maximum drawdown, Sharpe Ratio, Sortino Ratio, Mean-Variance optimization, Risk Parity, CVaR, and ARIMA and GARCH models. The practical part describes the developed application built using the Dash and Flask frameworks. The application enables transaction history import, management of multiple portfolios, reconstruction of historical portfolio value, calculation of performance and risk metrics, benchmark comparison, forecasting of future portfolio and individual asset development, and rebalancing proposals using three optimization approaches. An important part of the solution is the transformation of transaction data into time series and the use of a cash-flow-adjusted TWR index, which separates investment performance from external deposits and withdrawals. The result of the thesis is a functional analytical tool that connects statistical methods of financial analysis with practical decision support for an individual investor.