I'm creating a SPA using the backbone.js framework, the entire application is driven by a series of CSVs. The CSVs will look something like this.
Day, Time, Place, Score
Tuesday, 9:00 pm, Omaha, 13
Monday, 8:15 pm, KC, 15
The idea is that there will be a series of dropdown menus or even a drag and drop feature in which one can choose which of the csv headings to run a regression on. Obviously assigning dependent and independent variables.
I haven't been able to find a js library capable of doing a regression on the scale I want. The CSV will probably be around 300,000 rows. I'm relatively new to JS and am not looking to write this from scratch, if anyone has a method for OLS regression I would greatly appreciate it.
There is no existing statistical library that includes OLS Regression, but I found code for it here. http://trentrichardson.com/2010/04/06/compute-linear-regressions-in-javascript/
Here is the code:
function linearRegression(y,x){
var lr = {};
var n = y.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.length; i++) {
sum_x += x[i];
sum_y += y[i];
sum_xy += (x[i]*y[i]);
sum_xx += (x[i]*x[i]);
sum_yy += (y[i]*y[i]);
}
lr['slope'] = (n * sum_xy - sum_x * sum_y) / (n*sum_xx - sum_x * sum_x);
lr['intercept'] = (sum_y - lr.slope * sum_x)/n;
lr['r2'] = Math.pow((n*sum_xy - sum_x*sum_y)/Math.sqrt((n*sum_xx-sum_x*sum_x)*(n*sum_yy-sum_y*sum_y)),2);
return lr;
}
var known_y = [1, 2, 3, 4];
var known_x = [5.2, 5.7, 5.0, 4.2];
var lr = linearRegression(known_y, known_x);
// lr.slope
// lr.intercept
// lr.r2
This is easily the best one I found, and makes for an easy build into the Backbone.js framework.