Empirical results
Table III presents descriptive statistics[12] (mean, standard deviation) for the
explanatory variables defined in Table II, along with the results of a Kruskal-Wallis
test of means differences. The univariate tests suggest that the mean values of the
independent variables for the qualified versus the unqualified financial statements are
significantly different in almost all cases. More detailed, banks with qualified financial
statements appear to be smaller, more liquid, less well capitalized, and less cost
efficient on average than the ones with unqualified financial statements. We also
observe significant differences in the case of AUDRQ, DISCRQ, and OFDISPR.
Table IV presents the results of the multivariate analysis. Two specifications of the
logistic model are estimated[13]. The first specification (Model 1) includes only the five
bank specific variables (i.e. LOGASS, EQAS, ROAA, COST, LIQ), while the second
includes both the bank-specific variables and the country-specific variables (Model 2).
The results of the two estimations are reported in columns 1 and 2, respectively.
Both models are statistically significant at the 1 percent level with x 2 values equal
to 2,001.07 (Model 1) and 2,691.625, respectively. The Nagelkerke R 2 ¼ 0.47 (Model 1)
and 0.88 (Model 2), accordingly.
The logarithm of total assets (LOGASS) has a negative and statistically significant
coefficient in both cases indicating that the higher the size of the bank the lower the
probability of receiving a qualification. A possible explanation is that large companies
Empirical resultsTable III presents descriptive statistics[12] (mean, standard deviation) for theexplanatory variables defined in Table II, along with the results of a Kruskal-Wallistest of means differences. The univariate tests suggest that the mean values of theindependent variables for the qualified versus the unqualified financial statements aresignificantly different in almost all cases. More detailed, banks with qualified financialstatements appear to be smaller, more liquid, less well capitalized, and less costefficient on average than the ones with unqualified financial statements. We alsoobserve significant differences in the case of AUDRQ, DISCRQ, and OFDISPR.Table IV presents the results of the multivariate analysis. Two specifications of thelogistic model are estimated[13]. The first specification (Model 1) includes only the fivebank specific variables (i.e. LOGASS, EQAS, ROAA, COST, LIQ), while the secondincludes both the bank-specific variables and the country-specific variables (Model 2).The results of the two estimations are reported in columns 1 and 2, respectively.Both models are statistically significant at the 1 percent level with x 2 values equalto 2,001.07 (Model 1) and 2,691.625, respectively. The Nagelkerke R 2 ¼ 0.47 (Model 1)and 0.88 (Model 2), accordingly.The logarithm of total assets (LOGASS) has a negative and statistically significantcoefficient in both cases indicating that the higher the size of the bank the lower theprobability of receiving a qualification. A possible explanation is that large companies
การแปล กรุณารอสักครู่..

ผลการทดลองตารางที่สามนำเสนอสถิติเชิงพรรณนา [12] (เฉลี่ยส่วนเบี่ยงเบนมาตรฐาน) สำหรับตัวแปรที่กำหนดไว้ในตารางที่สองพร้อมกับผลของการKruskal-Wallis การทดสอบความแตกต่างของวิธีการ การทดสอบ univariate ชี้ให้เห็นว่าค่าเฉลี่ยของตัวแปรอิสระที่มีคุณสมบัติเหมาะสมสำหรับเมื่อเทียบกับงบการเงินอย่างไม่มีเงื่อนไขที่มีความแตกต่างกันในเกือบทุกกรณี รายละเอียดเพิ่มเติมธนาคารที่มีการเงินที่ผ่านการรับรองงบดูเหมือนจะเป็นขนาดเล็กสภาพคล่องมากขึ้นน้อยทุนที่ดีและค่าใช้จ่ายน้อยกว่าที่มีประสิทธิภาพโดยเฉลี่ยกว่าคนที่มีงบการเงินอย่างไม่มีเงื่อนไข Empirical results
Table III presents descriptive statistics[12] (mean, standard deviation) for the
explanatory variables defined in Table II, along with the results of a Kruskal-Wallis
test of means differences. The univariate tests suggest that the mean values of the
independent variables for the qualified versus the unqualified financial statements are
significantly different in almost all cases. More detailed, banks with qualified financial
statements appear to be smaller, more liquid, less well capitalized, and less cost
efficient on average than the ones with unqualified financial statements. We also
observe significant differences in the case of AUDRQ, DISCRQ, and OFDISPR.
Table IV presents the results of the multivariate analysis. Two specifications of the
logistic model are estimated[13]. The first specification (Model 1) includes only the five
bank specific variables (i.e. LOGASS, EQAS, ROAA, COST, LIQ), while the second
includes both the bank-specific variables and the country-specific variables (Model 2).
The results of the two estimations are reported in columns 1 and 2, respectively.
Both models are statistically significant at the 1 percent level with x 2 values equal
to 2,001.07 (Model 1) and 2,691.625, respectively. The Nagelkerke R 2 ¼ 0.47 (Model 1)
and 0.88 (Model 2), accordingly.
The logarithm of total assets (LOGASS) has a negative and statistically significant
coefficient in both cases indicating that the higher the size of the bank the lower the
probability of receiving a qualification. A possible explanation is that large companies
การแปล กรุณารอสักครู่..
