X hits on this document

28 views

0 shares

5 / 8

European Journal of Economics, Finance And Administrative Sciences - Issue 18 (2010) The estimation of inflation function using thr fixed effect model with weight (Cross Section

Weights) and white heteroscedasticity for the all region

s.

Variable

Coefficient

Standard Error

t-Statistic

Prob.

RER1?

0.046752

0.010840

4.312862

0.0000

YGAP1?

0.003611

0.011221

0.321826

0.7479

DPF?

0.457999

0.049634

9.227548

0.0000

DE?

0.020221

0.007389

2.736620

0.0067

DP1?

0.335427

0.102315

3.278372

0.0012

DA?*RER1?

-0.044973

0.011452

-3.926882

0.0001

DA?*YGAP1?

-0.005081

0.018135

-0.280176

0.7796

DA?*DPF?

-0.639722

0.153956

-4.155216

0.0000

DA?*DE?

-0.020993

0.013267

-1.582309

0.1149

DA?*DP1?

0.015914

0.148033

0.107503

0.9145

DC?

-0.007259

0.001471

-4.936225

0.0000

Weighted Statistics

R-squared

0.606783

Mean dependent var

0.558464

S.D. dependent var

S.E. of regression

0.023460

Sum squared resid

F-statistic

12.55785

Durbin-Watson stat

Prob(F-statistic)

0.000000

Unweighted Statistics

R-squared

0.568384

Mean dependent var

0.515346

S.D. dependent var

S.E. of regression

0.033290

Sum squared resid

Durbin-Watson stat

1.983705

0.051838 0.035306 0.129888 1.936611

0.035534 0.047819 0.261547

Note: RER1=Lag of real exchange rates (domestic currency/US\$); YGAP1=Lag output gap; DPF=Foreign inflation rate; DE=Nominal exchange rates change (domestic currency/US\$) and DP1=Lag of domestic inflation rate.

From the table, we learn that the coefficient of determination (R-Square) is 0.606783, which means more than 60% of the inflation variability inflation can be explained by the explanatory variables – output gap, nominal exchange rates and foreign as well as domestic inflation. The results also show that there are no multicollinearity and autocorrelation problems in the model. More detail analysis shows that inflation is significantly influenced by the lag of real exchange rates and domestic inflation, nominal exchange rates and foreign inflation. The output gap, however, have no significant impact on inflation. Coefficients of the model suggest that foreign and domestic inflation have the stronger impact in compare to real and nominal exchange rates. One percent increase in foreign inflation, for example, will be followed by about 0.46 percent increase in domestic inflation. One percent depreciation of exchange rates, on the other hand, will be followed by about 0.05 inflation rates.

Furthermore, the results also show that the behavior of inflation in Asian Region seems to be different with those of the European Union and North America. The coefficient of the interaction of area dummy variables with RER1 and DPF are statistically significant. Therefore, we further analyze and divide the model into two models, namely model for Asian Region and model for Non-Asian Region. The results are presented in Table 3 and Table 41.

1

We also find that the crisis dummy has also significant impact on the inflation. However, we could not establish the model before and after the Asian financial crisis due to the limited data availability.

 Document views 28 Page views 28 Page last viewed Fri Dec 09 13:44:41 UTC 2016 Pages 8 Paragraphs 380 Words 3347