How to regress OLS with the latest version of Pandas

Question

I wanted to run a maximized window of 1000 OLS regression estimationdataset for my assessment, found at the following URL:

https://drive.google.com/open?id=0B2Iv8dfU4fTUa3dPYW5tejA0bzg

I tried using the following Pythonscript with pandasversion 0.20.2.

# /usr/bin/python -tt

import numpy as np
import matplotlib.pyplot as plt
import pandas as pd
from statsmodels.formula.api import ols

df = pd.read_csv('estimated.csv', names=('x','y'))

model = pd.stats.ols.MovingOLS(y=df.Y, x=df[['y']], 
                               window_type='rolling', window=1000, intercept=True)
df['Y_hat'] = model.y_predict

However, when I run my Pythonscript, I get this error: AttributeError: module 'pandas.stats' has no attribute 'ols'. I found out the reason for this error, because it is deleted from pandasversion 0.20.0, as we see from the following link.

https://github.com/pandas-dev/pandas/pull/11898

How can we do OLS Regressionwith the latest version of Pandas?

Answer 1

- statsmodels.ols *, ( - 1000 258 . ), . , . . .

* statsmodels, . , . .

, yhat nx1 - nx1, , 1 1000- . , (257526, 1000), .

import numpy as np
import pandas as pd

df = pd.read_csv('input/estimated.csv', names=('x','y'))

def rolling_windows(a, window):
    """Creates rolling-window 'blocks' of length `window` from `a`.

    Note that the orientation of rows/columns follows that of pandas.

    Example
    =======
    onedim = np.arange(20)
    twodim = onedim.reshape((5,4))

    print(twodim)
    [[ 0  1  2  3]
     [ 4  5  6  7]
     [ 8  9 10 11]
     [12 13 14 15]
     [16 17 18 19]]

    print(rwindows(onedim, 3)[:5])
    [[0 1 2]
     [1 2 3]
     [2 3 4]
     [3 4 5]
     [4 5 6]]

    print(rwindows(twodim, 3)[:5])
    [[[ 0  1  2  3]
      [ 4  5  6  7]
      [ 8  9 10 11]]

     [[ 4  5  6  7]
      [ 8  9 10 11]
      [12 13 14 15]]

     [[ 8  9 10 11]
      [12 13 14 15]
      [16 17 18 19]]]
    """

    if isinstance(a, (Series, DataFrame)):
        a = a.values
    if a.ndim == 1:
        a = a.reshape(-1, 1)
    shape = (a.shape[0] - window + 1, window) + a.shape[1:]
    strides = (a.strides[0],) + a.strides
    windows = np.lib.stride_tricks.as_strided(a, shape=shape, strides=strides)
    return np.squeeze(windows)

def coefs(y, x):
    return np.dot(np.linalg.inv(np.dot(x.T, x)), np.dot(x.T, y))

rendog = rolling_windows(df.x.values, 1000)
rexog = rolling_windows(df.drop('x', axis=1).values, 1000)

preds = list()
for endog, exog in zip(rendog, rexog):
    pred = np.sum(coefs(endog, exog).T * exog, axis=1)
    preds.append(pred)
preds = np.array(preds)

print(preds.shape)
(257526, 1000)

: Random Forest Classifier , , y ?