|
Backtesting of Time Series Models
Stroukalová, Marika ; Houfková, Lucia (advisor) ; Zichová, Jitka (referee)
Title: Backtesting of Time Series Models Author: Marika Stroukalová Department: Department of Probability and Mathematical Statistics Supervisor: Mgr. Lucia Jarešová Supervisor's e-mail address: lucia.jaresova@centrum.cz Abstract: In the present work we study the basic models of financial time series (ARMA, GARCH), we focus on parameter estimation and forecasting in estimated models. We describe the means of estimating parametres and future values in the program R. In the theoretical section we also discuss the features of financial time series, define simple returns and log returns and we introduce the benefits of the log returns. We also apply the white noise model, ARMA(1,1) and GARCH(1,1) on historic time series of logarithmic returns of chosen stock exchange indices, we also backtest 1-step ahead fore- cats and 5-step ahead forecasts and we compare the results of these models. By empirical comparison of real data we also analyze how the models reac- ted on the present financial crisis and evaluate how the normal distribution assumption for the data held up. Keywords: time series, ARMA, GARCH, backtesting. 1
|
|
Prediction of Multiple Time Series at Stock Market Trading
Palček, Peter ; Zbořil, František (referee) ; Rozman, Jaroslav (advisor)
The diploma thesis comprises of a general approach used to predict the time series, their categorization, basic characteristics and basic statistical methods for their prediction. Neural networks are also mentioned and their categorization with regards to the suitability for prediction of time series. A program for the prediction of the progress of multiple time series in stock market is designed and implemented, and it's based on a model of flexible neuron tree, whose structure is optimized using immune programming and parameters using a modified version of simulated annealing or particle swarm optimization. Firstly, the program is tested on its ability to predict simple time series and then on its ability to predict multiple time series.
|
| |