Question

我正在尝试使用R中的poLCA包进行潜在类分析。我有284个变量（列）。逐个写入所有数据帧参数（列）是麻烦的并且是不可取的。我正在寻找一种简洁的方式来编写列而无需多打字。我已经生成了样本数据框，我用它来拟合一个简单的LCA模型：

library(poLCA);
set.seed(134);

#data

df<-as.data.frame(cbind(v1=rbinom(100, 1, 0.2)+1,
v2=rbinom(100,1,0.4)+1,
v3=rbinom(100,1,0.3)+1,
v4=rbinom(100, 1, 0.2)+1,
v5=rbinom(100,1,0.1)+1,
v6=rbinom(100,1,0.35)+1,
v7=rbinom(100, 1, 0.25)+1,
v8=rbinom(100,1,0.45)+1,
v9=rbinom(100,1,0.32)+1,
v10=rbinom(100, 1, 0.29)+1));

#poLCA model estimation - runs fine

f1<-cbind(v1,v2,v3,v4,v5,v6,v7,v8,v9,v10)~1;
system.time(fit<-poLCA(f1,data=df, nclass=3, maxiter=1000, graphs=FALSE,
tol=1e-10, na.rm=TRUE, probs.start=NULL));

#Instead of typing v1, v2, v3,.....,v10 in the formula 'f1' above, I re-wrote 'f1' 
as follows:

f2<-cbind(paste("v", c(1:10), sep=""))~1

#The above formula, 'f2', however combines all data frame arguments (columns) with
"double quote" wrapped around. Entering this formula inside poLCA function generates 
errors as "non-numeric arguments with mathematical function".  

#Thus, poLCA function below generates errors because f2 was not created correctly!!!

system.time(fit<-poLCA(f2, data=df, nclass=3, maxiter=1000, graphs=FALSE,
tol=1e-10, na.rm=TRUE, probs.start=NULL));

如何在上面创建f2等效的f1？我非常感谢您的回复。谢谢。

（系统信息：＆＃34; R版本3.1.0;＆＃34; x86_64-w64-mingw32＆＃34 ;;＆＃34; poLCA_1.4.1＆＃34;）

Answer 1

非常糟糕的解决方案。至少它有效，但我确信有更正确的方法。

f2 = as.formula(paste("cbind(", paste("v", c(1:10), sep = "", collapse = ","), ")~1", sep ="", collapse=""))

从字符串向量中删除双引号 - 在R中的poLCA应用程序

1 个答案: