PROSAGA码农传奇-R-forestplot使用不同的形状

<div class =“post-text”itemprop =“text”>
  
    
      我意识到我可以使用这个例子
      <a href="http://gforge.se/2013/12/the-forestplot-of-dreams/" rel="nofollow noreferrer">
        http://gforge.se/2013/12/the-forestplot-of-dreams/
      </A>
       制作两个单独的数据集，一个用于调整值，一个用于匹配值，然后组合 - 但不必再次输入所有值。
    
  
  
    我想你提到的链接（或
    <a href="https://cran.r-project.org/web/packages/forestplot/vignettes/forestplot.html#multiple-confidence-bands" rel="nofollow noreferrer">
      这个
    </A>
    ）是最佳解决方案。
您不必键入所有数据 - 只需在数据框中添加一个额外的列，然后对其进行子集化。您可以像这样添加列：
  
   <pre>
 <code>
 type = c(NA, NA, NA, "adjusted", "matched","adjusted", "matched", NA, "adjusted", "matched","adjusted", "matched", NA, "adjusted", "matched","adjusted", "matched", "adjusted", "matched")

</code>
 </pre>
  
    所以你的数据框将如下所示：
  
   <pre>
 <code>
 main_acevccb <- structure(list(
 mean = c(NA, NA, NA, -1.12, -0.64, -1.55,-1.60, NA, -1.35,-1.44, -1.3, -1.2, NA, -1.29,-1.23, -2.82,-2.15, -1.84,-2.72), 
 lower = c(NA, NA, NA, -1.41, -0.84, -1.85, -1.86, NA, -1.71,-1.9, -1.57,-1.52, NA, -1.53, -1.54, -4.04, -3.61, -2.85,-4.45),
 upper = c(NA, NA, NA, -0.83, -0.44, -1.26, -1.34, NA, -1.0, -0.98,-1.04, -0.87, NA, -1.04,-0.93, -1.59,-0.68, -0.82, -0.99),
 type = c(NA, NA, NA, "adjusted", "matched","adjusted", "matched", NA, "adjusted", "matched","adjusted", "matched", NA, "adjusted", "matched","adjusted", "matched", "adjusted", "matched")),
 .Names = c("Difference", "lower", "upper", "type"), 
 row.names = c(NA, -19L), 
 class = "data.frame")

</code>
 </pre>
  
    ......以及它的子集：
  
   <pre>
 <code>
 adjusted <- subset(main_acevccb, type!="matched"|is.na(type))
matched <- subset(main_acevccb, type!="adjusted"|is.na(type))

</code>
 </pre>
  
    然后，您将有两个单独的数据框用于调整和匹配的值，并且可以按照链接中的描述继续进行。
  
</DIV>