Home > Search Results: LOJDA, Patrik
Search:
Search Tips :: Advanced Search
Search collections:
Sort by: Display results: Output format:
National Repository of Grey Literature 2 records found  Search took 0.01 seconds. 
Portfolio optimization
 LOJDA, Patrik
This diploma thesis is focused on portfolio optimization for a selected client. As a first task, a set of 35 company shares are chosen out of 6 business areas in order to ensure variety. With the help of Markowitz theory, optimal portfolios are created. In the beginning, it is mainly about optimization tasks that focus on only one of the two conflicting goals. Followed by modifications of the optimization function and of course putting those portfolios in a graph. By adding a risk-free asset we create a capital market line that achieves better results than the original Markowitz theory. The thesis also uses a security market line for the evaluation of individual shares, from which the optimal portfolio is compiled. Fama-French three-factor model is also presented. The resulting portfolios are tested on the next year data and with the help of the acquired knowledge an optimal portfolio is created for the selected client.
Detailed record
Statistical analysis of unemployment
 LOJDA, Patrik
My bachelor thesis is focusing on creation of an unemployment rate model based on various characteristics regarding all districts in the Czech Republic. Beginning of this thesis is dedicated to labor market, income and substitution effect which both affect the job offer. Afterwards we uncover various types of unemployment, including detailed description of statistical methods used inside and description of correlation and regression analysis which are important parts for next steps. Main point of the during model creation is definitely choosing variables. We try to find suitable ones. Correlation analysis and graphs are used for checking the validity of all variables. This way 3 models are created while using linear regression method in Statistica 13.2. Models are made of data considering specific years. We use them for 2016 unemployment rate prediction and compare final results. Second point of this thesis is focused on area correlation. We discovered that the 2005 year model is totally useless for predication. The 2015 model seems to be fine, but results are on higher average than true values. The 2010 year model is the best out of those three, but isn´t suitable for accurate predictions. After analysis of all models several findings have been found. Variables "distance from Prague" and "household IT equipment" are present in all models. We can say that unemployment rate is generally lower in districts closer to Prague and in households with lower values of IT equipment. We found a correlation of unemployment rate in neighborhood, but for it to be taken as a fact we should do deeper analysis.
Detailed record

Interested in being notified about new results for this query?
Subscribe to the RSS feed.
Digital Repository :: Search :: Submit :: Personalize :: Help
Powered by Invenio v1.1.2
Maintained by admin.invenio@techlib.cz
This site is also available in the following languages:
Česky  English