Dirk Eddelbuettel — Dec 21, 2012 | source

This example was motivated by http://stackoverflow.com/questions/5158219/find-minimum-of-vector-in-rcpp and addresses how to find the minumum value and its position index in a vector.

```
#include <Rcpp.h>
using namespace Rcpp;
// [[Rcpp::export]]
double vecmin(NumericVector x) {
// Rcpp supports STL-style iterators
NumericVector::iterator it = std::min_element(x.begin(), x.end());
// we want the value so dereference
return *it;
}
```

We can also use the iterator to compute the position, simply by taking the offset to the vector beginning.

```
// [[Rcpp::export]]
int vecminInd(NumericVector x) {
// Rcpp supports STL-style iterators
NumericVector::iterator it = std::min_element(x.begin(), x.end());
// we want the value so dereference
return it - x.begin();
}
```

A quick illustration follows. Note that we pad the position by one to adjust for the 0-based versus 1-based indexing between C++ and R.

```
set.seed(5)
x <- sample(1:100, 10) # ten out 100
x
```

[1] 21 68 90 28 11 67 50 76 88 96

```
cat("Min is ", vecmin(x), " and at position ", vecminInd(x)+1, "\n")
```

Min is 11 and at position 5

Of course, we subsequently added `min`

and `which_min`

as sugar
functions, but this example still illustrated how useful the STL
algoritms can be.

**tags:**
stl

- Call Python from R through Rcpp — Wush Wu
- Using iterators for sparse vectors and matrices — Soren Hojsgaard and Doug Bates
- A simple array class with specialized linear algebra routines — Fabian Scheipl