Added instrument data collection with tests

Author: keeganskeate

README.md

@@ -16,6 +16,10 @@ You can install the Cannlytics engine from [PyPI](https://pypi.org/project/cannl
 pip install cannlytics
 ```
 
+## Documentation
+
+Please refer to the [Cannlytics Developer Guide](https://docs.cannlytics.com/) for detailed information about the module.
+
 ## Authentication, Data, and File Management
 
 Cannlytics leverages Firebase by default for a database, file storage, and authentication.

cannlytics/analytics/var.py

@@ -0,0 +1,95 @@
+"""
+VAR Functions | Cannlytics
+
+Author: Keegan Skeate
+Contact: <keegan@cannlytics.com>
+Created: Wed Apr 14 07:55:43 2021
+License: MIT License <https://opensource.org/licenses/MIT>
+
+Description:
+
+    Crude VAR functions.
+
+Resources:
+
+    https://www.statsmodels.org/stable/generated/statsmodels.tsa.tsatools.lagmat.html
+
+"""
+
+import numpy as np
+import pandas as pd
+import statsmodels.api as sm
+from statsmodels.tsa import tsatools
+
+
+def VAR(Vector, lag_order):
+    """
+    Inputs a Vector of dimension N x I, where N is the number of observations
+    and I is the number of variables, as well as the lag order of the model.
+    Returns the estimated equations from OLS as a dictionary with names 'Eq#'.
+    """
+    X = np.empty_like(Vector)
+    for i in range(1 , 1 + lag_order):
+        X = np.column_stack([X, lag(Vector, i)])
+    X = np.delete(X, np.s_[0:len(Vector[0])], axis=1)
+    X = sm.add_constant(X)   
+    VAR_estimates = {}
+    for i in range(1,len(Vector[0])+1):
+        VAR_estimates["Eq{0}".format(i)] = sm.OLS(Vector[:,[i-1]][lag_order:],
+                                                  X[lag_order:]).fit()       
+    return VAR_estimates
+
+    
+def VAR_forecast(Vector, VAR_estimates, lag_order, horizon,shock=None):
+    """
+    Inputs the VAR Vector, VAR estimates, the lag order of the model,
+    the forecast horizon, and the desired first period shock.
+    """
+    # Inital Period shock for IRF
+    if shock is None:
+        shock = np.zeros(len(Vector[0]))
+    error = np.zeros((len(Vector),len(Vector[0])))
+    error[0] = shock
+    # Predictions for Forecast Horizon
+    for t in np.arange(0,horizon):     
+        X_hat = Vector
+        for i in range(1 , lag_order):
+            # X_hat = np.column_stack([X_hat, lag(Vector, i)])
+            lagged_vector = tsatools.lagmat(Vector, maxlag=1)
+            X_hat = np.column_stack([X_hat, lagged_vector])
+        X_hat = sm.add_constant(X_hat)     
+        Y_hat = []
+        for Eq in VAR_estimates:
+            Y_hat.append(np.dot(X_hat[-1], VAR_estimates[Eq].params))
+        Forecast = Y_hat + error[t]
+        Vector = np.vstack((Vector,Forecast))
+    return Vector[-horizon:]
+
+
+def IRF(Vector,VAR_estimates,lag_order,horizon,shock):
+    baseline = VAR_forecast(Vector,VAR_estimates,lag_order,horizon,shock=None)
+    impact = VAR_forecast(Vector,VAR_estimates,lag_order,horizon,shock=shock.T)
+    return (impact-baseline)
+
+
+def lag_series(series, lag=None):
+    lagged_series = pd.Series(series).shift(1)
+    return lagged_series.values
+
+
+def lag(x,lag=None):
+    if lag==None: lag=1
+    lag_values = np.empty_like(x)
+    for i in np.arange(0,len(x)):
+        if i>=lag:
+            lag_values[i] = x[i-lag]
+    return lag_values
+
+
+def cov_matrix(u):
+    k = len(u[0])
+    matrix = np.ones((k,k))
+    for i in np.arange(0,k):
+        for j in np.arange(0,k):
+            matrix[i][j] = np.cov(u.T[i],u.T[j])[0][1]
+    return matrix