基于R语言 数据检验详解
<div id="navCategory"><h5 class="catalogue">目录</h5><ul class="first_class_ul"><li>1. W检验（Shapiro–Wilk (夏皮罗–威克尔 ) W统计量检验)</li><li>2. K检验(经验分布的Kolmogorov-Smirnov检验)</li><li>3. 相关性检验：</li><li>4. T检验</li><li>5. 正态总体方差检验</li><li>6. 二项分布总体假设检验</li><li>7. Pearson 拟合优度χ2检验</li><li>8. Fisher精确的独立检验：</li><li>9. McNemar检验：</li><li>10. 秩相关检验</li><li>11. Wilcoxon秩检验</li></ul></div><p class="maodian"></p><h3>1. W检验（Shapiro–Wilk (夏皮罗–威克尔 ) W统计量检验)</h3><blockquote><p>目标：检验数据是否符合某正态分布，如：标准正态分布N(0,1)<br />R函数：shapiro.test().<br />结果含义：当p值小于某个显著性水平α(比如0.05)时，则认为样本不是来自正态分布的总体，否则认为样本来自正态分布的总体。</p></blockquote>
<p class="maodian"></p><h3>2. K检验(经验分布的Kolmogorov-Smirnov检验)</h3>
<blockquote><p>目标：检验数据的分布是否符合函数F(x)<br />R函数:ks.test(),如果P值很小，说明拒绝原假设，表明数据不符合F(n,m)分布。</p></blockquote>
<p class="maodian"></p><h3>3. 相关性检验：</h3>
<div class="jb51code"><pre class="brush:plain;">R函数：cor.test()</pre></div>
<div class="jb51code"><pre class="brush:plain;">cor.test(x, y,
alternative = c("two.sided", "less", "greater"),
method = c("pearson", "kendall", "spearman"),
exact = NULL, conf.level = 0.95, ...)</pre></div>
<blockquote><p>结果含义：如果p值很小，则拒绝原假设，认为x,y是相关的。否则认为是不相关的。</p></blockquote>
<p class="maodian"></p><h3>4. T检验</h3>
<blockquote><p>目标：用于正态总体均值假设检验，单样本，双样本都可以。 <br />R函数：t.test()</p></blockquote>
<p style="text-align:center"><img alt="在这里插入图片描述" src="https://img.jbzj.com/file_images/article/202203/2022030211181034.png" /></p>
<div class="jb51code"><pre class="brush:plain;">t.test(x, y = NULL,
alternative = c("two.sided", "less", "greater"),
mu = 0, paired = FALSE, var.equal = FALSE,
conf.level = 0.95, ...)</pre></div>
<blockquote><p>结果意义：P值小于显著性水平时拒绝原假设，否则，接受原假设。具体的假设要看所选择的是双边假设还是单边假设（又分小于和大于）</p></blockquote>
<p class="maodian"></p><h3>5. 正态总体方差检验</h3>
<div class="jb51code"><pre class="brush:plain;">R函数：t.test()</pre></div>
<div class="jb51code"><pre class="brush:plain;">t.test(x, y = NULL,
alternative = c("two.sided", "less", "greater"),
mu = 0, paired = FALSE, var.equal = FALSE,
conf.level = 0.95, ...)</pre></div>
<blockquote><p>结果意义：P值小于显著性水平时拒绝原假设，否则，接受原假设。具体的假设要看所选择的是双边假设还是单边假设（又分小于和大于）</p></blockquote>
<p style="text-align:center"><img alt="在这里插入图片描述" src="https://img.jbzj.com/file_images/article/202203/2022030211181035.png" /></p>
<p class="maodian"></p><h3>6. 二项分布总体假设检验</h3>
<div class="jb51code"><pre class="brush:plain;">binom.test(x, n, p = 0.5,
alternative = c("two.sided", "less", "greater"),
conf.level = 0.95)</pre></div>
<p>原假设：p=p0,p<p0,p<p0 计算结果p-值很小，表示拒绝假设，否则为接受假设.</p>
<p class="maodian"></p><h3>7. Pearson 拟合优度χ2检验</h3>
<div class="jb51code"><pre class="brush:plain;">chisq.test(x, y = NULL, correct = TRUE,
p = rep(1/length(x), length(x)), rescale.p = FALSE,
simulate.p.value = FALSE, B = 2000) </pre></div>
<p>原假设H0：X符合F分布。</p>
<p class="maodian"></p><h3>8. Fisher精确的独立检验：</h3>
<div class="jb51code"><pre class="brush:py;">fisher.test(x, y = NULL, workspace = 200000, hybrid = FALSE,
control = list(), or = 1, alternative = "two.sided",
conf.int = TRUE, conf.level = 0.95)</pre></div>
<p>原假设：X,Y相关。</p>
<p class="maodian"></p><h3>9. McNemar检验：</h3>
<div class="jb51code"><pre class="brush:plain;">mcnemar.test(x, y = NULL, correct = TRUE)</pre></div>
<p>原假设：两组数据的频数没有区别。</p>
<p class="maodian"></p><h3>10. 秩相关检验</h3>
<div class="jb51code"><pre class="brush:plain;">cor.test(x, y,
alternative = c("two.sided", "less", "greater"),
method = "spearman", conf.level = 0.95, ...)</pre></div>
<p>原假设：x,y相关.</p>
<p class="maodian"></p><h3>11. Wilcoxon秩检验</h3>
<div class="jb51code"><pre class="brush:plain;">wilcox.test(x, y = NULL,
alternative = c("two.sided", "less", "greater"),
mu = 0, paired = FALSE, exact = NULL, correct = TRUE,
conf.int = FALSE, conf.level = 0.95, ...)</pre></div>
<p>原假设：中位数大于，小于，不等于mu</p>
<p>到此这篇关于R语言 数据检验的文章就介绍到这了,更多相关R语言 数据检验内容请搜索琼殿技术社区以前的文章或继续浏览下面的相关文章希望大家以后多多支持琼殿技术社区！</p>
<div class="art_xg">
<b>您可能感兴趣的文章:</b><ul><li>R语言批量读取某路径下文件内容的方法</li><li>R语言因子类型的实现</li><li>R语言向量下标和子集的使用</li><li>R语言使用cgdsr包获取TCGA数据示例详解</li></ul>
</div>
</div>
<!--endmain-->
頁:
[1]