The relationship between macroeconomic indicators and economic development based on time-series model in the case of Australia Boqijonov Adxamjon Ikromjon o’g’li
Table 1: Data of economic indicators of Australia in 1981-2021
Download 233.91 Kb.
|
Australia model (2)
- Bu sahifa navigatsiya:
- Exchange rate Industry
- Table 2: The results of GDP per capita based on Dickey-Fuller
- Table 3: The result of Inflation based on Dickey-Fuller
- Table 4: The result of FDI based on Dickey-Fuller
- Table 5: The result of unemployment based on Dickey-Fuller
- Table 6: The result of export based on Dickey-Fuller
- Table 7: The result of exchange rate based on Dickey-Fuller
- Table 8: The result of regression of economic indicators
- Table 9: The result of correlation of economic indicators
- Table 10: The result of regression of logarithmic economic indicators
- Table 11: Gaus Markovs second condition
- Figure 1. Scatter plot Table 12: the Breusch-Pagan test result
- Table 13: The White test result
- Table 14: Durbin-Watson test result
- Table 15: Breusch-Godfrey test result
- Table 16: Shapiro-Wilk test results
Table 1: Data of economic indicators of Australia in 1981-2021
Table 2: The results of GDP per capita based on Dickey-Fuller From table 2 we can identify that the result of Dickey-Fuller test for the GDP per capita. We did two integrations to make data stationary and our second integration was the stationary statistical test value in this table. Table 2 shows that 1 percent critical value is -3.655, the 5 percent critical value is –2,961, and the 10 percent critical value is - 2,613. they are greater than the statistical test value and it proves that this indicator is stationary now and the p-value should be smaller that 0.05 and we could achieve that result, which indicates a strong stationary presence with a small value. Table 3: The result of Inflation based on Dickey-Fuller In table 3, we can see that the result of Dickey-Fuller test for the inflation. We did also two integrations to find the result of Dickey-Fuller test for the inflation. The statistical test value is -4.264 which is smaller than other critical values. 1% percent critical value is -3,655, the 5 percent critical value is –2,961, the 10 percent critical value is - 2,613. P-value is 0.0000, which indicates a strong stationary presence Table 4: The result of FDI based on Dickey-Fuller From table 4, we can see the result of Dickey-Fuller test for FDI. We did one integration in this section and we achieved the supposed result which shows The statistical test value is –6.592 which is much smaller than other critical values. 1% percent critical value is -3,648, the 5 percent critical value is –2,958, the 10 percent critical value is - 2,612. P-value is 0.0000 which is less than 0.05. Table 5: The result of unemployment based on Dickey-Fuller Table 5 shows us the result of Dickey-Fuller test for unemployment. We did also two integrations to find the result of Dickey-Fuller test for unemployment and we got the supposed result. We can see from that table our result is stationary test value. Table 6: The result of export based on Dickey-Fuller Table 6 shows that the result of Dickey-Fuller test for export. The statistical test value is -4.540 which is smaller than other critical values. 1% percent critical value is -3.655, the 5 percent critical value is –2.961 the 10 percent critical value is - 2,613. P-value is 0.0000, which indicates a strong stationary presence. Table 7: The result of exchange rate based on Dickey-Fuller Table 7 shows the result of exchange rate based in Dickey-Fuller. In this part we achieved the supposed result by doing two integrations. The statistical test value is -4.423 which is smaller than other critical values. 1% percent critical value is -3.655, the 5 percent critical value is –2.961 the 10 percent critical value is - 2,613. P-value is 0.0003, which indicates a strong stationary presence. Table 8: The result of regression of economic indicators In table 8, we can the regression table which is a statistical method which shows the strength of the relationship between a dependent variable and one or more independent variables. It is used to predict whether how 1 unit change in one variable effects to another variable Table 9: The result of correlation of economic indicators This table shows the calculation and the description of the correlation coefficient. The correlation coefficient describes how one variable moves in relation to another. The correlation coefficient is the specific measure that quantifies the strength of the linear relationship between two variables in a correlation analysis. Table 10: The result of regression of logarithmic economic indicators 𝑙𝑛𝐺𝐷𝑃𝑝𝑒𝑟𝑐𝑎𝑝𝑖𝑡𝑎i = 6.17 – 0,17 ln Unemployment + 0,20lnExporti + 𝜀i Economic indicators such as unemployment, export, inflation, industry, and other model indicators have a long-term link with economic development (GDP per capita), as seen by the correlation between the aforementioned indicators and GDP per capita. According to this regression coefficient, a 1% change in foreign direct investment causes a 0.000005% increase in GDP per capita, a 1% change in inflation causes a 0,007% decrease in GDP per capita, a 1% change in unemployment causes a 0.17% decrease in GDP per capita, and a 1% change in exports causes a 0.20% increase in GDP per capita. We continue our diagnostic examination of the econometric model we built in light of the widely accepted Gauss-Markov criteria. The first criterion of Gauss-Markov states that there should be six times as many observations as indicators in an econometric model. With forty observations and six indicators, we can say that the first Gauss-Markov criterion has been met by our model. Gauss Markov's second criterion enables us to observe that the econometrical model we developed is almost equal to the whole amount of theoretical data we acquired from the World Bank, which is shown in table 11 below. Table 11: Gaus Markov's second condition We can conclude table 11 that our model successfully passed through Gaus Markov's second condition. The third requirement requires us to determine if the residue is connected to the model or not. We may state that the econometric model has passed the third criteria if the residual is not connected to it. A heteroskedastic state is one in which the residues and the model are connected. To examine the connection between the model and the residue, we have three options. These are the correlation table, the test method, and the graph method. The relationship between the model and the residue for the test method was examined using the White test and the Breusch-Pagan test. Figure 1. Scatter plot Table 12: the Breusch-Pagan test result The Breusch-Pagan test results show that the residues are not connected to the model because the p value is greater than 0.05, which is referred to as the homoscedastic state by this test criterion. The alternative hypothesis is accepted since the zero hypothesis demonstrates that there is no heteroskedasticity in the residuals. In other words, the structured model's residuals exhibit a homoscedastic oscillation. The next step is to run the White test on the econometric model. This test also needs the p value to be higher than 0.05, just like the Breusch-Pagan test above. Table 13: The White test result If we look at Table 13's p-value, the White test's p-value is higher than 0.05, rejecting the heteroscedasticity criterion that the White test calls for. As a result, we can now accept alternative hypothesis 1. We must test our model against the fourth criterion, which is that the model's residuals should not have an autocorrelation issue. Table 14: Durbin-Watson test result Durbin-Watson statistics will always return a value between 0 and 4. A value of 2.0 implies that no autocorrelation was discovered in the data. Values ranging from 0 to less than 2 indicate positive autocorrelation, whereas values ranging from 2 to 4 indicate negative autocorrelation. When we ran this test on the model, the result was 1.032361, indicating that the residuals were correlated and that our model did not pass the Durbin-Watson test. The Breusch-Godfrey test will be used next to see if there are any autocorrelation issues in the residuals. Table 15: Breusch-Godfrey test result As seen in table 15, if chi2 is less than 0.05 or 5%, the null hypothesis can be rejected. In other words, the residuals in the model have a serial correlation. As a result, adjust for the assumption of no serial correlation. Table 16: Shapiro-Wilk test results When we examine the Gauss-Markov's fifth condition, the Shapiro-Wilk value is greater than p>0.05. The model can meet the requirements of the fifth condition, according to the information in the Shapiro-Wilk test. Figure 2: Normal distribution of residuals test Figure 2 demonstrates that the distribution of the residues is typical. VIF will be seen in the following phase. Table 17: VIF test There is no longer any variable in the model that has the problem of multicollinearity. The sixth criteria of the Gaus Markov technique was met by our model. All variable indications are less than ten. Table 18: Var Model The following is the VAR model specification: where is the intercept, a constant, and the coefficients of the delays of Y up to order p. The order 'p' indicates that up to p-lags of Y are utilized as predictors in the equation. The _t represents the error, which is regarded as white noise. Table 18 demonstrates that the independent variable, inflation, has a negative impact on the dependent variable at lag 1, and the link is not statistically significant at two lags. As a result of lags 1 and 2, we can conclude that there is a negative correlation between inflation and GDP per capita. Based on the data in our VAR regression table, we created the VAR model formula shown below. Yt =3218 + 0.557L1 GDP per capita t-1 - 0.415L2 GDP per capita t-2 – 183.08L1 Inflationt-1 + 21.41 L2 Inflationt-2 -537.17 L1 Unemploymentt-1 +599.93 L2 Unemploymentt-2 +εt Based on the results of our VAR model, we will project GDP per capita from 2022 to 2026 in the following phase. We employed the VAR and ARDL models in our investigation, and it was discovered that the VAR model was more effective in putting our study's prognosis in a good position, and as a consequence, we met our research goal. Figure 3: Forecast In conclusion, the GDP per capita in the Australia may rise in the following five years. We can predict that the Netherlands' GDP per capita will be higher than it was in 2021. We can see from this prediction that Inflation rate decrease over the next five years in Australia. And there is a fluctuation in Unemployment rate in the next five year period. DISCUSSION We used a multi-factor time series, namely a VAR model, to test our hypothesis on the effects of unemployment, FDI, export, an industry, and inflation on GDP per capita, which measures a country's economic progress. Then we validated our findings by running our model through the VIF test, which it failed. As a result, we found the problematic variable in the model, eliminated it, and retested the new model using Gaus Markov's five requirements, hoping that it satisfied them all. We also started determining whether or not each of the Australia's economic indices and leftovers were stationary. After inspecting, we determined that all indications needed to be rendered stationary, which we accomplished after two integrations. We opted to employ the VAR model to obtain forecasting-relevant findings once the new model passed the identification and evaluation tests. Looking at the graph above, we can see that the new model gave 5 projections for the Australia's future GDP per capita levels. CONCLUSION The new econometric model demonstrates that many economic indicators have both positive and negative effects on a country's GDP per capita, which measures economic stability. These include low productivity, high structural unemployment, climate change mitigation, population aging, and inequality. To enhance the Australian economy, they should expand exports, stimulate industrial production, and maintain a low inflation rate. The country's economy is ranked highly in the world. However, the country's companies and banks must become more internationalized through exports and direct investment abroad. Author Contributions Conceptualization: Boqijonov Adhamjon Data curation: Alitoshev Diyorbek Formal analysis: Abdurahmonov Ozodbek Investigation: Boqijonov Adhamjon Methodology: Abdurahmonov Ozodbek Software: Alitoshev Diyorbek Supervision: Boqijonov Adhamjon Visualization: Alitoshev Diyorbek Writing – original draft: Abdurahmonov Ozodbek REFERANCES World Bank: https://data.worldbank.org/ https://www.atlantis-press.com/article/125982626.pdf https://www.tcmb.gov.tr/wps/wcm/connect/aa8abd7e-3165-4900-8b2d-ea2238640dcb/15Inflation.pdf?MOD=AJPERES&CACHEID=ROOTWORKSPACE-aa8abd7e-3165-4900-8b2d-ea2238640dcb-m3fxB6Z Download 233.91 Kb. Do'stlaringiz bilan baham: 
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||