Chapter Four Empirical Results................................ 51
4.1 Unit Root Test and Co- integration Test Results ............................... 51
4.2 Vector Error Correction Model and Granger Causality test. ................ 53
Chapter Five Conclusions ............... 57
5.1. Major Findings. ................................ 57
5.2 Results Discussion and Policy Implications.................... 59
Chapter Four Empirical Results.
4.1 Unit Root Test and Co- integration Test Results
The past conducted researches and scholars emphasize that most time series variables arenon-stationary and using non-stationary variables in a model might lead to spurious regressions.We use quarterly data starting from the first quarter of 1997 to last quarter of 2018, which wascollected from World Bank official website (2019). Unit root test was employed to determine thetime series characteristic of the variables. As we can see, some of the variables show non-stationarydata, which means, are unpredictable and cannot be modeled or forecasted. The results obtainedby using non-stationary time series may be spurious in that they may indicate a relationshipbetween two variables where one does not exist.
In hypothesis testing on ADF, a critical value is a point on the test distribution that iscompared to the test statistic to determine whether to reject the null hypothesis or not. EViewsshows the critical values at the 1%, % 5 and 10% levels. If the absolute value of the test statisticis greater than the critical value, you can declare statistical significance and reject the nullhypothesis. In our test results, statistic value is less than critical value, which means we fail toreject null hypothesis. We run Phillips - Perron test to find out whether all variables are integratedof same order or not.
Table 4-1. AUGMENTED DICKEY FULLER (ADF) AND PHILLIPS-PERRON (PP)TESTS FOR STATIONARITY
...........................
Chapter Five Conclusions
5.1. Major Findings.
Phillips-Perron test was carried for stationarity to check if all the variables are integrated ofthe same order. Making sure they are, Johansen Co-integration test was implied and foundpresence of co-integration between foreign direct investment and dependent variables. VECM wasour next choice for using with non-stationary series that are known to be co-integrated. Thevariables get more than a co-integrating vector, so a VECM that can be applied to both short andlong term variables is an effective estimation method. Following, Granger Causality concludedwith final results and we had strong evidence to reject null hypothesizes in almost all variablesexcept Exchange Rate.
While the GDP impacts are of primary importance here, the three other economic variableshave been included as "control variables" for checking the excessively high estimates for the FDI.The final results matched the prior expectations, except Exchange Rate’s position towards FDIwas completely different than expected (no effect). The outcomes of the experiments indicate thatGDP has a favorable and statistically relevant impact on FDI. This means that, for the timeprescribed for this investigation (1997-2018), the increase in GDP in the economy of Uzbekistandemonstrated that FDI was a contributing factor to the country's economic development.
According to the foregoing discussion, it should be pointed out that although the governmentof Uzbekistan have been making very reasonable efforts in attracting FDI, particular economicand political circumstances existed in the country have hindered investment inflow and its overallperformance. Let it be noted that between 2011 and 2015, foreign investment in Uzbekistan hadreduced from $1.6 billion to $634.5mln, a drop of 60 percent (the biggest drop happened in theeconomy after the Independence). Most of the investors are concerned about legalization of theirbusiness and regulations that put the investment into risk.
reference(omitted)
