《秩检验国际会计专题4pa课件.ppt》由会员分享，可在线阅读，更多相关《秩检验国际会计专题4pa课件.ppt（14页珍藏版）》请在三一文库上搜索。
1、秩检验国际会计专题4pa,1,Structure of Date Analysis & Results,Index of comparability-net incomeIndex of comparability-shareholders equityFrequency table of distribution of values of index comparabilityIndex of comparability by industryIndex of comparability by size,Results of Wilcoxon testResults of Krusk
2、al-Wallis test,Frequency of the adjustmentsPartial index of comparabilityResults of kruskal-wallis test for partial index,秩检验国际会计专题4pa,2,Table-6 Frequency table of distribution of values of index of comparability,Table 6 classifies all observations in terms of their level of materiality.,With regard
3、 to overall income, a large proportion of the observations (64.3%)were material (differences exceeding 10%).,However,45.35% of material observations related to a decrease in net income under US GAAP, which confirms their relevance in increasing the mean index.,秩检验国际会计专题4pa,3,Table-6 Frequency table
4、of distribution of values of index of comparability,Index value,以表6可比性指数的频率分布做出如下柱状图：,直观来看，大于或小于1的数值的频率分布相差较多，呈现出这些年来拉丁美洲国家的可比性指数的平均值和中位值都显著地不同于1.0。但这些直观的结论是不足的，下文还有进一步的统计学证据。,秩检验国际会计专题4pa,4,Table 7-Index of comparability by industry,After considering factors like country and year, the results for t
5、he other factors reveal that the maximum restatement of net income by industry was carried out by companies included in non-financial entity(index 1.34).However, the maximum variation of shareholders equity was carried out by companies included as financial entities(index 2.14), adding adjustments t
6、o the domestic figure.,秩检验国际会计专题4pa,5,Table 8- Index of comparability by size,The medium-sized companies were the ones that modified their domestic net income to a greater absolute extent to calculate it according to US GAAP(index 1,95).However ,the largest-sized companies were the ones that modifie
7、d their domestic shareholders equity to a greater absolute extent in order to calculate it according to US GAAP(index 1.51).,秩检验国际会计专题4pa,6,From the table 4,5,7,8,we analyzed four different factors including country, year, industry and size, which may affect values of index of comparability. We find
8、 that almost all means or median values of index of comparability are above 1.0. The results from the analysis of overall differences in net income and shareholders equity between Latin American GAAP and US GAAP indicate that Latin American practices were less conservative than US GAAP during the pe
9、riod, in terms of the impact on fundamental accounting variables.However, dose this preliminary conclusion is supported by statistical evidence?Then comes to our non-parametric test.,Preliminary conclusion,秩检验国际会计专题4pa,7,Background knowledge for Non-parametric test,Parametric tests have requirements
10、 about the nature or shape of the populations involved. Nonparametric tests do not require that samples come from populations with normal distributions or have any other particular distributions. Consequently, nonparametric tests are called distribution-free tests.关于参数检验和非参数检验方法的选择：凡是符合或经过变换后符合参数检验条
11、件的资料，最好用参数检验。当资料不具备参数检验的条件时，非参数检验是一种有效的分析方法。,秩检验国际会计专题4pa,8,Disadvantages of Nonparametric Methods 1,tend to waste information 2, are not so efficient as parametric tests, so we need stronger evidence before we reject a null hypothesis.,Advantages of Nonparametric Methods 1,can be applied to a wide
12、variety of situations. 2,can be applied to categorical or nominal level data 3,involve simpler computations and are easier to understand and apply.,在现实的研究中，我们遇到的数据常常具备以下特点:( 1) 数据的总体分布类型未知; 或( 2) 数据的总体分布类型已知，但不符合正态分布; 或( 3) 某些变量可能无法精确测量非参数方法适用于：非正态分布的资料，方差不齐的资料，等级资料，一端或两端有不确定性数值（如）1.1，.0.9等）的资料，分布不明
13、的资料。例如，本文的研究样本。,秩检验国际会计专题4pa,9,威尔科克森符号秩检验,For exampleAs usual, we will use the five step hypothesis-testing procedure.H0:There is no difference in the ratings of the two flavors H1:The spicy ratings are higherThis is a one tailed test. The significance level is 0.05.The steps to conduct the Wilcoxon
14、 signed rank test are as follows. Step (1) - Compute the difference between the spicy flavor score and the current flavor score for each participant. Step (2) - Only the positive and negative differences are considered further. Step (3) - Determine the absolute differences for the values computed in
15、 column 4. Step (4) - Rank the absolute differences from smallest to largest. Step (5) - Each assigned rank in column 6 is then given the same sign as the original difference, and the results are reported in column 7and 8. Step (6) -The R+ and R- columns are totaled.,秩检验国际会计专题4pa,10,秩检验国际会计专题4pa,11,
16、The critical values for the Wilcoxon signed-rank test are located in Appendix. The row is used for one-tailed tests and the 2 row for two-tailed tests. In this case we want to show that customers like the spicy taste better, which is a one-tailed test, so we select the row. We chose the 0.05 signifi
17、cance level, so move to the right to the column headed 0.05. Go down that column to the row where n is 14. Then we get it. The value at the intersection is 25, so the critical value is 25.The decision rule is to reject the null hypothesis if the smaller of the ranksums is 25 or less. In this case th
18、e smaller rank sum is 30, so the decision isnot to reject the null hypothesis.We cannot conclude there is a difference in the flavor ratings between thecurrent and the spicy. Mr Peterson has not shown that customers preferthe new flavor.,类似地，将本例对威尔科克森符号秩检验的应用运用到本文的研究中去，即研究拉丁美洲公司报告的基本会计变量是否显著高于或低于美国报
19、告的净收益和股东权益。,秩检验国际会计专题4pa,12,Table 9 indicates that statistically significantdifferences between Latin American and US net income occurred in the years 1999,2000 and 2001 (at p0.05). Moreover,It confirms the significance of Argentinas, Chiles and Mexicosresults. There were also significant difference
20、s for industry, and small and medium-sizedcompanies.,The results for the shareholders equity index confirm that theindex value for all the analyzed factors was significantly greater than 1.,秩检验国际会计专题4pa,13,Because the computed value of H(5.736) is not beyond 5.991, the null hypothesis is not rejecte
21、d. There is no evidence of a difference among the executives from accounting, finance, and engineering with respect totheir typical knowledge of economics principles.,类似地，将本例对Kruskal-Wallis秩和检验的应用运用到本文的研究中去，即研究年份、国家、行业和规模各个因素是否分别对可比性指数的值有重要影响。,秩检验国际会计专题4pa,14,The results from the Kruskal-Wallis test
22、 conclude that there were no statistically significant relationships between country, year, company size or industry and the quantitative variation of the domestic net income to be restated as US GAAP.,Regarding the quantitative variation of domestic shareholders equity, all the analyzed factors, except year, influenced the index of comparability.,