inv cumulative probability for t-distribution

package.json

@@ -1,8 +1,11 @@
 {
   "name": "js-stats",
   "version": "1.0.0",
-  "description": "Package implements various statistic methods and distribution functions such as normal, fisher and student-t, and so on",
+  "description": "Package implements various statistics and distribution function",
   "author": "Xianshun Chen",
+  "contributors": [
+    "Xianshun Chen <xs0040@gmail.com>"
+  ],
   "license": "MIT",
   "main": "index.js",
   "directories": {

src/jsstats.js

@@ -89,6 +89,8 @@ var jsstats = jsstats || {};
 
 		this.ntable = [ 1.880794, 1.959964, 2.053749, 2.170090, 2.326348, 2.575829 ];
 
+		this.percentiles = [0.970, 0.975, 0.980, 0.985, 0.990, 0.995];
+
 		this.ttable = 
         [
             [10.57889, 12.7062, 15.89454, 21.20495, 31.82052, 63.65674],
@@ -256,6 +258,18 @@ var jsstats = jsstats || {};
     	return tcdf;
 	};
 
+	TDistribution.prototype.invCumulativeProbability = function(p, df) {
+		if(!df){
+			df = this.df;
+		}
+		var delta = 0.005;
+        var row = df - 1;
+        var column = Math.round((p - this.percentiles[0]) / delta);
+        column = (column >= 0) ? column : 0;
+        column = (column < this.percentiles.Length) ? column : this.percentiles.length - 1;
+        return (row < this.ttable.length) ? this.ttable[row][column] : this.ntable[column];
+    };
+
 	jsstats.TDistribution = TDistribution;
 
 

test/t-distribution-spec.js

@@ -15,6 +15,10 @@ describe("Create t distribution", function() {
       var distribution = new jsstats.TDistribution(10.0);
       expect(distribution.cumulativeProbability(0.0)).to.equal(0.5);
     });
+    it('has t_df = 0 with probability of 0.5', function(){
+      var distribution = new jsstats.TDistribution(10);
+      expect(distribution.invCumulativeProbability(0.5)).to.equal(0.0);
+    });
   });