The determinants of local government performance have long been of interest, and have been empirically investigated in a number of recent studies. A general question in this area is: why have some local governments across the countries achieved high rates of performance whilst the others remained at lower levels? Evidence from the relevant literature indicates that some local governments or municipalities, particularly in developed countries, achieve very good performance. Others, in particular those from underdeveloped or developing, have very little or almost no performance.

This research empirically investigates the budgeting conditions in North Cyprus and evaluate how budgeting conditions vary from one municipality to the other in the five major municipalities; Lefkosa, Magusa, Girne, Guzelyurt and Iskele over the period 2004Q1 to 2012Q4. Two types of methodologies are conducted. First; it empirically investigates the determinants of local government deficit by using cointegration techniques. The Johansen co-integration and The Vector Error Correction model (VECM) are applied to test the effect of each element such as own-source revenue, grants, expenditures and debt services on the budget deficit. Second, the study test the convergence hypothesis if similar regions within the country converge to the same long-run steady state by considering five major Cypriot municipalities in the North Cyprus economy.

The results from the empirical analysis on budget deficit findings show that there is evidence of a long-run relationship between general deficit and its determinants. The evidence of causality is found from expenditure and debt to general deficit for Lefkosa municipality. There is unidirectional causality from grant and revenue through deficit for Magusa municipality. Third empirical analysis on convergence introduced that the municipalities with the lowest number of people per staff, commonly diverge from the group mean.

Understanding the behavior of asset prices is essential in evaluating investment decisions so that practitioners and academics have been investigating what assets are worth and whether asset prices are predictable. The degree of stock return predictability is a statistical concept and totally an empirical issue. In this study, we conduct an empirical application on the equity premium forecastability concept. We consider both macroeconomic and technical indicators in order to predict the equity risk premiums of selected thirteen stock markets including Belgium, Greece, Malaysia, Mexico, Portugal, Spain, Taiwan, Turkey, Brazil, Hong-Kong, India, Russia, and South Africa. Incorporating information from all indicators including domestic and foreign macroeconomic indicators, oil price changes, and technical indicators provides in-sample fitting gains. Out-of-sample forecasting results show that the strategy of combining forecasts provides out-of-sample gains. Portfolio performance analyses suggests following the forecasts generated by the proposed models to obtain positive returns over the buy-and-hold strategy.