python - Python loop - stochastic vs batch loop, and why its taking too long?

Question

Loop 1: (Stochastic - expected faster)

 for i in range(len(data)):
        x = data[i]
        for t in range(0, m):
            theta[t] -= my_lambda * (np.dot(theta, x) - y[i]) * x[t]

vs Loop 2: (Batch - expected slower)

for t in range(0, m):
        temp = 0
        for i in range(len(data)):
            x = data[i]
            temp += (np.dot(theta, x) - y[i]) * x[t]
        theta[t] = theta[t] - my_lambda * temp / M

But my batch is much much much faster - One iteration Elapsed time 0.025356054306 vs my stochastic is really slower - One iteration Elapsed time 1.0576338768

Here my m is 13 and data is around 500 tuples. Can anyone tell me why so? Or where did I mess up for stochastic? Thanks!

EDIT 1:

data = [[], [], ...]
data = [record1, record2 ...]
record = [1, x1, x2, ... , x_m]

theta is vector
theta = np.zeros(m+1) //np is numpy