import numpy as np
import os
from sklearn.naive_bayes import GaussianNB
from helper_package import helper_functions as hf
from helper_package import feature_set as fs
from helper_package import confusion_matrix_calcs as cm
from helper_package import assign_labels as al
# raises numpy errors/warnings so they can be caught by try/except
np.seterr(all='raise')
# allow df console output to display more columns
hf.show_more_df()
# get DataFrame of stock ticker info from csv file
df = hf.fix_column_names(hf.get_ticker_df())
df = al.assign_color_labels(df) # assign color labels
df = fs.get_feature_set(df) # add mean and std return columns for DF
def nb_predict(df1, df2):
"""
Gaussian Naive Bayesian classification of labels (colors)
:param df1: Training set (DataFrame)
:param df2: Prediction set (DataFrame)
:return: df2 with predicted label (binary) and color columns
"""
try:
x = df1[['Mean_Return', 'Std_Return']].values
transactions in year 2?
6. Are these results very different from those in year 1 for this value of W?
"""
import pandas as pd
import numpy as np
import os
import copy
import matplotlib.pyplot as plt
from sklearn.linear_model import LinearRegression
from helper_package import helper_functions as hf
import Assignment4.window_strategy as ws
hf.show_more_df() # allow df output to display more rows/columns
df = hf.fix_column_names(hf.get_ticker_df()) # get DataFrame of ticket file
def create_w_list(w_min, w_max, step=1):
"""
Create list of w values (used as 'windows' for estimation)
:param w_min: Minimum w value
:param w_max: maximum w value
:param step: step for w values in list
:return output_list: list of w values
"""
output_list = [] # list of w values
for i in range(w_min, w_max + 1):
output_list.append(i)
i += step