econkit 0.4.1.9.9.4

Advanced Econometric Analysis Tools

EconKit - Function Descriptions

econkit is a Python library that provides various statistical and econometric analysis tools, including descriptive statistics, correlation matrices, and tests for stationarity and autocorrelation.

Functions

---

Descriptive Statistics

descriptives(data)

Computes descriptive statistics for each numeric column in a DataFrame.

Parameters:

• data: pandas.DataFrame containing the data to be analyzed.

Returns:

• None. Prints a summary table of the descriptive statistics.

Example Usage:

import pandas as pd
from econkit import econometrics as ec

df = pd.read_csv('your_data.csv')
ec.descriptives(df)

---

Correlation Matrix

correlation(df, method='Pearson', p=False)

Calculates and prints the correlation matrix and p-values for numeric columns in the provided DataFrame. Supports Pearson, Spearman, and Kendall correlation methods.

Parameters:

• df: pandas.DataFrame containing the data to be analyzed.
• method: str (optional). Method of correlation ('Pearson', 'Spearman', or 'Kendall'). Default is 'Pearson'.
• p: bool (optional). If True, p-value matrix is also printed; if False, only the correlation matrix is printed.

Returns:

• None. Prints the correlation matrix and optionally the p-value matrix.

Example Usage:

import pandas as pd
from econkit import econometrics as ec

df = pd.read_csv('your_data.csv')
ec.correlation(df, method='Spearman', p=True)

---

Augmented Dickey-Fuller (ADF) Test

adf(dataframe, maxlag=None, regression='c', autolag='AIC', handle_na='drop')

Performs the ADF test on each column in the DataFrame and returns a summary table.

Parameters:

• dataframe: pandas.DataFrame containing the data to be tested.
• maxlag: int (optional). Maximum number of lags to use.
• regression: str (optional). Type of regression trend ('c', 'ct', 'ctt', 'nc'). Default is 'c'.
• autolag: str (optional). Method for lag length selection ('AIC', 'BIC', 't-stat').
• handle_na: str (optional). How to handle missing values ('drop' or 'fill').

Returns:

• None. Prints a summary table of the ADF test results.

Example Usage:

import pandas as pd
from econkit import econometrics as ec

df = pd.read_csv('your_data.csv')
ec.adf(df)

---

KPSS Test

kpss(dataframe, regression='c', nlags='auto', handle_na='drop')

Performs the KPSS test on each column in the DataFrame and returns a summary table.

Parameters:

• dataframe: pandas.DataFrame containing the data to be tested.
• regression: str (optional). Type of regression trend ('c' or 'ct'). Default is 'c'.
• nlags: str or int (optional). Number of lags to use. Default is 'auto'.
• handle_na: str (optional). How to handle missing values ('drop' or 'fill').

Returns:

• None. Prints a summary table of the KPSS test results.

Example Usage:

import pandas as pd
from econkit import econometrics as ec

df = pd.read_csv('your_data.csv')
ec.kpss(df)

---

Durbin-Watson Test

dw(data)

Performs the Durbin-Watson autocorrelation test and Ljung-Box test for each column of the dataset.

Parameters:

• data: pandas.DataFrame containing the time series data.

Returns:

• None. Prints a summary table of the Durbin-Watson test results.

Example Usage:

import pandas as pd
from econkit import econometrics as ec

df = pd.read_csv('your_data.csv')
ec.dw(df)

---

Normality Tests

normality_tests(data, pvalue=False)

Performs normality tests (Shapiro-Wilk and Kolmogorov-Smirnov) on each column of a DataFrame and displays results as a formatted table.

Parameters:

• data: pandas.DataFrame containing the numerical data.
• pvalue: bool (optional). If True, displays the p-value matrix along with the test results. Default is False.

Returns:

• None. Prints the results matrix and optionally the p-value matrix.

Example Usage:

import pandas as pd
from econkit import econometrics as ec

df = pd.read_csv('your_data.csv')
ec.normality_tests(df, pvalue=True)

---

Random Number Generator

random_number_generator(num_vars, num_random, distribution, mean=0, std_dev=1, seed=None, lower=None, upper=None, p=None, n_trials=None, lambda_param=None, start=None, stop=None, step=None, repeat_each=None, repeat_sequence=None)

Generates random numbers based on the specified distribution and parameters.

Parameters:

• num_vars: int. Number of variables (columns) to generate.
• num_random: int. Number of random values per variable.
• distribution: str. The distribution type ('normal', 'uniform', 'binomial', 'bernoulli', 'poisson', 'patterned').
• mean: float. Mean for the normal distribution. Default is 0.
• std_dev: float. Standard deviation for the normal distribution. Default is 1.
• lower: float. Lower bound for the uniform distribution.
• upper: float. Upper bound for the uniform distribution.
• p: float. Probability for Bernoulli or Binomial distributions.
• n_trials: int. Number of trials for the Binomial distribution.
• lambda_param: float. Lambda parameter for the Poisson distribution.
• start: float. Start value for the patterned distribution.
• stop: float. Stop value for the patterned distribution.
• step: float. Step size for the patterned distribution.
• repeat_each: int. Number of times to repeat each number in the patterned distribution.
• repeat_sequence: int. Number of times to repeat the entire sequence.

Returns:

• pandas.DataFrame. A DataFrame containing the generated random numbers.

Example Usage:

from econkit import econometrics as ec

random_data = ec.random_number_generator(num_vars=2, num_random=10, distribution='normal', mean=0, std_dev=1, seed=42)
print(random_data)

---

Combine DataFrames

combine_dataframes(column_name, *dataframes)

Combines multiple DataFrames based on a specified column into a new DataFrame, aligning by the index.

Parameters:

• column_name: str. The column to extract from each DataFrame.
• *dataframes: Multiple pandas DataFrames passed as positional arguments.

Returns:

• pandas.DataFrame. A new DataFrame combining the specified columns from all input DataFrames.

Example Usage:

import pandas as pd
from econkit import econometrics as ec

df1 = pd.read_csv('data1.csv')
df2 = pd.read_csv('data2.csv')

combined_df = ec.combine_dataframes('Close', df1, df2)
print(combined_df)

---

Financial Data Retrieval

data(ticker_symbol, start_date, end_date, interval)

Downloads financial data from Yahoo Finance and calculates daily returns.

Parameters:

• ticker_symbol: str. The stock ticker symbol.
• start_date: str. Start date in 'dd-mm-yyyy' format.
• end_date: str. End date in 'dd-mm-yyyy' format.
• interval: str. Data interval (e.g., '1d', '1wk', '1mo').

Returns:

• pandas.DataFrame containing the stock data and calculated returns.

Example Usage:

from econkit import finance as f

start = '01-06-2024'
end = '07-06-2024'
interval = '1d'

df = f.data('AAPL', start, end, interval)
print(df.head())

---

Usage Notes

• Ensure your data is clean and properly formatted before using these functions.
• Some functions handle missing values; specify your preferred method using the handle_na parameter.
• For time series analysis, ensure your data is indexed by date.

For further details, refer to the function docstrings in the source code or the examples provided above.