error - Multiple Regression OLS class

Question

I having problem to adjust with ols class the equation y = b₀ + b₁x¹ + b₂x²

The code:

xs = numpy.loadtxt('teste.csv', skiprows=1, dtype=float, delimiter=';', usecols=(0,1)) y = log(xs[:,0]) x = 1/xs[:,1] x2 = (1/xs[:,1])**2 mymodel = ols.ols(y,x,'y',['x1', 'x2']) mymodel.summary()`

I got this error:

print '''% -5s % -5.6f % -5.6f % -5.6f % -5.6f''' % tuple([self.x_varnm[i],self.b[i],self.se[i],self.t[i],self.p[i]]) IndexError: index out of bounds

Someone can help me?

Jaime · Accepted Answer · 2013-04-03 21:01:29Z

1

Try defining your x as:

x = 1/xs[:,1:2] # slice to keep (n, 1) shape x2 = (1/xs[:,1:2])**2 x = np.hstack((x, x2))

You are telling ols to expect a two column matrix for x, but are passing a single one in, hence the error.

answered Apr 3, 2013 at 21:01

Jaime

67.7k19 gold badges128 silver badges164 bronze badges

Sign up to request clarification or add additional context in comments.

1 Comment

askewchan Over a year ago

or x = 1/xs[:,1,np.newaxis] which looks less like a mistake to me. x[1:2] is the kind of thing I'll go back and change in old code :P

Collectives™ on Stack Overflow

error - Multiple Regression OLS class

1 Answer 1

1 Comment

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

1 Comment

Related