Descriptive analysis of this study is presented in table-1, related to mean, maximum, minimum, standard deviation, skewness, kurtosis, Jarque-Bera
and observations of endogenous and exogenous variables.
145.5795 802.7583 552.4852 Skewness 0.157839 0.289803 0.244415 Kurtosis 1.598340 2.039601 2.122258 Jarque-Bera
219.9342 134.0623 107.5417 Probability 0.000000 0.000000 0.000000 Sum 3268659.
Pelo teste Jarque-Bera
, de aderencia a distribuicao normal dos dados, apenas as variaveis Indicadores, Receita e Ranking apresentaram distribuicao normal (p>0,05), impossibilitando testes parametricos em todos os casos.
statistics indicates that the series under analysis are non-normal for the crisis periods.
To confirm the results of Normality and Heteroscedasticity tests, Jarque-Bera
and Breusch-Pegan-Godfrey tests were applied successfully.
The normality of residual errors assessed by Jarque-Bera
test indicated problems, as expected; since data are spatially correlated.
5079.5140 0.9799 2.0445 14.0749 Skewness 0.7622 -1.2255 -1.0842 0.0873 Kurtosis 4.2591 9.3493 5.5461 2.2499 Jarque-Bera
18.2416 (***) 216.1639 (***) 52.1965(***) 2.7681 Analysis LIR Mean 2.0904 Maximum 7.47467 Minimum -0.1450 Std.
test of both dataset rejects the hypothesis of normality.
Amongst others, these tests include the Kolmogorov-Smirnov (KS) test , the Lilliefors (LL) test , the Anderson-Darling (AD) test [3, 4], the Shapiro-Wilks (SW) test , the Jarque-Bera
(JB) test , and the DAgostino and Pearson (DP) test .
Em primeiro lugar, na Tabela 4, o resultado do teste Jarque-Bera
(JB) nao rejeita a hipotese de normalidade dos residuos.
The value of Jarque-Bera
was 2003 with significant p-value suggested that the problem of normality existed in the data but since the sample size was far beyond 30, according to the central limit theorem, should the sample size be above 30, data is asymptotically normal.
Both variables do not present normal distribution by the Jarque-Bera
test, as presented in Table 1.