A project Undertaken as Part of bsc (Hons) Economics with Finance degree, Westminster International University in Tashkent

Results

Unit root test

The important thing about OLS regression is to check the data against OLS assumptions. One of the essential assumption is stationary data. Consequently, the ADF and PP unit root tests are carried out to identify if data obeys this assumption.
Descriptive statistics and graphical illustration of the collected data must be analyzed before doing any econometric tests.
The below graphs illustrate visual representation of data for the given period.


The table below shows the descriptive statistics of variables during the period of 1974-2017.

Table 1. Descriptive statistics

Variable	Observations	Mean	Std. Dev	Min	Max
regbus	36	25.90222	9.528054	13.08	39.35
lnroi	28	.0654425	.4418197	-.6018089	.7470741
dgdp	43	-.0577404	4.936342	-14.68216	13.49703
lnCPI	44	.9853589	.7773616	-1.237847	2.852381
drate	30	.0084954	5.543367	-13.15107	15.92273
tf_ratio	36	2.485984	.7417478	1.437604	3.830463

The table above shows descriptive statistics of variables that have been chosen for the model. As we can see from the table, number of observations are different in variables. The highest number of observation belongs to lnCPI and the least amount is 28, which is lnroi’s observation. The reason variables such as lnroi, dgdp, lnCPI and drate have negative values is due to they are not expressed in nominal values, but are first differentials and logarithm values.
Mostly the mean in observations are less than one. This is because all of the variable proxies are expressed in either logarithm, differences or ratios. Only regbus and tf_ratio has different amount, the mean for regbus and tf_ratio is 25,902,220 and 2.485984 respectively. Standard Deviation of the variables is also shown in the table. Minimum and maximum values are also provided in the table. The maximum registered businesses during these years was 39.35 million and minimum amount is 13.08 million. GDP was 13.5 percent at its highest level and the lowest level was -14.7 percent and on average -0.58 percent. The maximum and minimum values of inflation was 2.85 and -1.24 percent respectively.
Table 2 shows the Augmented Dickey-Fuller (ADF) and Philips-Perron (PP) tests’ results. The null hypothesis for this unit root tests is that the variables have unit root that is non-stationary and alternative hypothesis is that the variable is stationary.