The aim of the study is to examine the impact of public expenditure on economic growth of Kosovo. Time series data span for the period of time 2002-2015. The structure of the econometric model is built on Keynesian theories and endogenous growth model. The model estimation is performed only after implementing the Augmented Dickey-Fuller (ADF) Unit Root test to estimate if time series are stationary. Several tests have been implemented to determine model validity. The model has met all the assumptions of statistical tests: error term residuals have a normal distribution (Jarque-Bera test), there is no auto-correlation between variables (Breusch-Godfrey Serial test), and error variances are constant, known as the principle of homoscedasticity (Breusch-Pagan-Godfrey test). Gross domestic product is used as a dependent variable in the model, while public expenditure (G), foreign direct investment (FDI), export (EXP) and total budget revenue (TrTax) are used as the endogenous variables. The study results have revealed that there is a positive and statistically significant effect of public expenditures and exports on economic growth. Total budget revenue has a positive impact on economic growth but this has not been proved to be statistically significant. The authors of the research have also found out that FDI is negative and statistically insignificant.