Simple linear regression in Javascript

linear_regression.js

function linear_regression(x_array, y_array){
        var slope;
        var intercept;

        var n = y_array.length;
        var sum_x = 0;
        var sum_y = 0;
        var sum_xy = 0;
        var sum_xx = 0;
        var sum_yy = 0;

        for (var i = 0; i < y_array.length; i++) {
            sum_x += x_array[i];
            sum_y += y_array[i];
            sum_xy += (x_array[i] * y_array[i]);
            sum_xx += (x_array[i] * x_array[i]);
            sum_yy += (y_array[i] * y_array[i]);
        } 

        slope = (n * sum_xy - sum_x * sum_y) / (n * sum_xx - sum_x * sum_x);
        intercept = (sum_y - slope * sum_x) / n;

        return [ slope, intercept ];
}

/** demo **/
var weights = linear_regression([1,2,3], [24, 36, 49]);

var x = 4;
var prediction = x * weights[0] + weights[1];