Search

In this study, it is aimed to analyze the factors that may stimulate international tourism demand for Turkey. In empirical analyses, the fuzzy logic approach to gravity model with data from 83 countries that sent tourists to Turkey in 2019 are used. GDP per capita (GDPpp), population (POP), distance (DIST), inflation (INF), exchange rate (EXCR), tourism expenditure (EXPD) and tourist departure (DEP) variables are used in the analysis. As a result of the study, only population, distance, expenditure and departure variables were found to be significant and only population variable was found to be fuzzy. According to this finding, the effect of the population variable on the number of tourists appears to be positive in all cases. However, the unit effect of change in population is not clear-cut and can be expressed by values varying within a range. [ABSTRACT FROM AUTHOR]

In this study, the purpose is to develop an alternative approach reconsidering on the definitions of the expected return and the risk in Modern Portfolio Theory. In this alternative approach, the value at the end of investment period of each asset is forecasted by using ARIMA models. Percentage differences between forecasted values and present values are accepted as the expected returns. Risk value is obtained by calculating where the present value of the asset is located in the confidence interval of forecasted value. Then, optimization is carried out by employing Quadratic Programming. Finally, performances of created portfolios are compared to the results of Modern Portfolio Theory. The alternative approach seems more effective than traditional one. [ABSTRACT FROM AUTHOR]

In this paper, a simulated annealig (SA) algorithm developed to solve resource constrained multi-project problem is presented. Object of programming is to minimize project delays plus standby times of activities. From this perspective, this study differs from similar studies. The feasible solution representation is based on a precedence feasible list of activities and resurce assignments. In this study unlike traditional SA algorithms , a very slow cooling schedule and two different stop criteria are used. The algorithm is tested on a randomly generated large problem. The results confirm the effectiveness of the proposed algorithm. [ABSTRACT FROM AUTHOR]