Correlation and Regression Lines - A quick recap#3

Solution Hackerrank

Correlation and Regression Lines - A quick recap #3.py

@@ -0,0 +1,44 @@
+
+#RAMESH CHANDRA
+
+import math
+x = [15 , 12 , 8  , 8 ,  7 ,  7 ,  7 ,  6   , 5  , 3]
+y = [10  ,25 , 17  ,11 , 13 , 17 , 20 , 13 , 9  , 15]
+#calculate mean of x and y
+sum_x,sum_y=0,0
+for i in xrange(len(x)):
+    sum_x,sum_y=sum_x+x[i],sum_y+y[i]
+
+mean_x = float(sum_x)/len(x)
+mean_y = float(sum_y)/len(y)
+#now calculate r = A/sqrt(B*C)
+# A = sum ((x[i]-x_mean_x)*(y[i]-mean_y)) 
+# B = sum(x[i]-mean_x)^2
+# C = sum(y[i]-mean_y)^2
+
+A, B , C ,= 0,0,0;
+for i in xrange(len(x)):
+    A += (x[i]-mean_x)*(y[i]-mean_y)
+    B += (x[i]-mean_x)**2
+    C += (y[i]-mean_y)**2
+#Co-relation Coefficient
+r = float(A)/math.sqrt(B*C)
+#slope  b = r*(sd_x/sd_y)
+#s_d and s_y are standard deviations of x and y respectively
+std_x = math.sqrt( (float(B)/(len(x))))
+std_y = math.sqrt( (float(C)/(len(y))))
+#final slop
+b = float(r*std_y)/std_x
+
+#regression line
+# y = a+ bx
+
+# a = mean_y - b*mean_x
+a = mean_y - b*mean_x
+
+#regression line
+# y = a+ bx
+ans = a+b*10
+print("%.1f"%ans)
+
+#for more info....go to this link   https://www.youtube.com/watch?v=GhrxgbQnEEU