Weighted mean

Optimizing R Code with Rcpp

Romain François

Consulting Datactive, ThinkR

Weighted mean of x with weights w

R version

# see also ?weighted.mean
weighted_mean_R <- function(x, w){
    sum(x*w) / sum(w)
}

R version

# see also ?weighted.mean
weighted_mean_R <- function(x, w){
    sum(x*w) / sum(w)
}

R version

# see also ?weighted.mean
weighted_mean_R <- function(x, w){
    sum(x*w) / sum(w)
}

R version

# see also ?weighted.mean
weighted_mean_R <- function(x, w){
    sum(x*w) / sum(w)
}

Inefficient R version

weighted_mean_loop <- function(x, w){
    total_xw <- 0
    total_w  <- 0

    for( i in seq_along(x)){
        total_xw <- total_xw + x[i]*w[i]
        total_w  <- total_w  + w[i]
    }

    total_xw / total_w

}

Skeleton of a C++ version

// [[Rcpp::export]]
double weighted_mean_cpp( NumericVector x, NumericVector w){
    double total_xw = 0.0 ;
    double total_w  = 0.0 ;

    int n = ___ ;

    for( ___ ; ___ ; ___ ){
        // accumulate into total_xw and total_w
    }

    return total_xw / total_w ;

}

Missing values

Testing if a value is a missing value in a numeric vector

bool test = NumericVector::is_na(x) ;

The representation of NA in double

double y = NumericVector::get_na() ;

Let's practice!

Optimizing R Code with Rcpp