PCA reports error in eigen (crossprod (t (x), t (x)), symmetric = true): 'x' has infinite value or missing value

How about missing values PCA

> dat.pca <- PCA(dat, graph = FALSE)
Error in eigen(crossprod(t(X), t(X)), symmetric = TRUE) : 
  'x' There are infinite values or missing values in 
In addition: Warning message:
In PCA(dat, graph = FALSE) :
  Missing values are imputed by the mean of the variable: you should use the imputePCA function of the missMDA package

But come up with the following tips , Say let me use it imputaPCA, I know this after searching imputePCA yes missMDA A function in a package , So you need to download missMDA package , And put the PCA Switch to imputePCA. So what is missMDA? As shown in the figure below , It's a kind of matrix There are missing values , But this package can be used for these missing value pairs PCA The resulting image is not large , This function imputePCA That is to say, those with missing values matrix These missing values can become an estimable value , Then you can still use PCA Medium FactoMineR To calculate . It's fine too Reference resources
Husson and Josse Wrote a book called missMDA My bag , Sum up PCA Analyze all possible methods of interpolating missing values through iteration .imputePCA（） Function can interpolate missing values . Please check the help of this function .

Josse, J., Husson, F.: Handling missing values in exploratory multivariate data analysis methods.
Journal de la Société Française de Statistique. 153, 79–99 (2012)

PCA Missing value