找回密码
 立即注册

QQ登录

只需一步,快速开始

搜索

SPSS卡方的拟合优度检验,检验一列数据是否为随机分布,怎么做?

[复制链接]
发表于 2016-8-28 18:07:35
打扰到您了,陈老师。我现在有一列观测值的分布,我的目的比较简单,就是检验其是否为随机分布。我在文献上找到了类似的操作步骤,内容如下:The numbers of recombination breakpoints present in each gene region and in incremented windows of sizes 200 and 500 nucleotides were counted. A Chi-squared goodness of fit test was used to see whether the observed numbers of recombination breakpoints differ significantly from a random distribution. The expected numbers of breakpoints for each window were assumed to follow a Poisson discrete distribution. The appropriateness of this distribution was tested by simulating the locations of breakpoints randomly (10,000 times) and for each performing the Chi-squared goodness of fit test. The number of times the Chi-squared value for the simulated data exceeded the value for the Poisson distributed was counted.  
他说的比较详细,但我外行不太懂,想请您帮忙解决。谢谢您!




使用道具 举报 回复
发表于 2016-8-28 20:50:45
你查到的资料是卡方拟合优度检验,卡方拟合优度是检验一列变量是否服从某一指定的分布,具体的讲解见如下视频:http://v.youku.com/v_show/id_XMTI2NDEwNDQzNg==.html?from=s1.8-1-1.2&spm=0.0.0.0.5NQVg0

然而检验一列变量的分布是否是随机的,选择的统计学方法是...成为会员可查看全部,【点击成为会员】
使用道具 举报 回复 支持 反对
发表于 2016-8-29 18:14:53
陈老师,您好,感谢您的回复和指点,您讲的很清楚。但我还有几个较大疑惑麻烦您解答:1,视频例子当中男女随机分布是什么意思?是说沿着X轴上每一次抽样,取Y轴上男女(0或1)时都是随机的吗?还只是说,男女最后总的数值相当,所以可见是随机取的?2,我查询资料得知,游程检验的必要条件是变量是二类取值,且变量个数有限制,那么对于我的数据(近一百个数,而且每个数是0到100的连续变量)通过勾选中位数和均值得出的结果会不会降低检验效能,能不能满足国际高水平期刊的统计要求?还有没有其他检验随机分布的方法以便两者互相应证?3,我们贴给您的英文文献中的方法针对的问题和我们的完全一致,也是检验和我们同类数据的是否随机分布,您能大概给我们讲解一下其思路吗?麻烦您了,以后有问题终于有可靠的老师请教了,期待和您的长期合作,从今天开始我就是您的忠实粉思啦!
使用道具 举报 回复 支持 反对
发表于 2016-8-30 12:18:01
陈老师 发表于 2016-8-28 20:50
你查到的资料是卡方拟合优度检验,卡方拟合优度是检验一列变量是否服从某一指定的分布,具体的讲解见如下视 ...

The numbers of recombination breakpoints present in each gene region(对每一个基因的表达和不表达是否随机,选择游程检验) and in incremented windows of sizes 200 and 500 nucleotides(对于incremented windows,表达的计数服从Poisson分布 were counted. A Chi-squared goodness of fit test was used to see whether the observed numbers of recombination breakpoints differ significantly from a random distribution(卡方拟合优度检验是检验breakpoints显著不是随机分布!. The expected numbers of breakpoints for each window were assumed to follow a Poisson discrete distribution(每一个window的期望计数expected number,服从泊松分布,那么就显著不是随机分布). The appropriateness of this distribution was tested by simulating the locations of breakpoints randomly (10,000 times) and for each performing the Chi-squared goodness of fit test. The number of times the Chi-squared value for the simulated data exceeded the value for the Poisson distributed was counted.(对incremented windows 的表达次数进行记录,记录出每一种次数出现的频次,并运用Poisson分布的公式计算出每一种次数的期望频次,最后进行卡方拟合优度检验,看实际频次与期望频次是否拟合,拟合则说明数据服从泊松分布,而非随机分布

通过卡方拟合优度进行泊松分布的检验比较麻烦,可以直接在...成为会员可查看全部,【点击成为会员】
使用道具 举报 回复 支持 反对
您需要登录后才可以回帖 登录 | 立即注册

本版积分规则

 
QQ在线咨询