Vectorised way to calculate mean of left and right neighbours in a vector

Question

I have a vector

x = c(1820.0, 2235.0, 2534.0, 2580.0, 2322.0, 2317.0, 2331.0, 2345.0, 
      2305.0, 2265.0, 2277.0, 2289.0, 2338.0, 2387.0, 2152.0, 2256.0, 
      2360.0, 2590.0, 2529.0, 2468.0, 2776.0, 2909.0, 3017.0, 3081.0,
      3118.5, 3156.0, 3338.0, 3211.5)

I want to calculate the mean of left and right neighbors of each element except edges. For example, the result should be like this : mean(1820,2534), mean(2235,2580), mean(2534,2322) ...

I am able to do this using loops but that is very slow. I need a vectorized solution.

My code using a for loop:

neighbour_m = function(x) {
  newx = c(x[length(x)], x, x[1])
  for (i in 2:(length(newx) - 1)){
    m = mean(c(newx[i-1], newx[i+1]))
  }
}

Answer 1

Another base R option using rowMeans and cbind

rowMeans(cbind(x[1:(length(x) - 2)], x[3:(length(x))]))
# [1] 2177.00 2407.50 2428.00 2448.50 2326.50 2331.00 2318.00 2305.00 2291.00
#[10] 2277.00 2307.50 2338.00 2245.00 2321.50 2256.00 2423.00 2444.50 2529.00
#[19] 2652.50 2688.50 2896.50 2995.00 3067.75 3118.50 3228.25 3183.75