Shock and scenario analysis

Scenario Analysis and Shock Implementation in MARIO¶

Implementing a shock without using mario excel platform¶

A scenario can be implemented in multiple ways within mario. Simple shocks may be implemented with mario as follow:

In [1]:

from mario import load_test, slicer

In [2]:

# loading the baseline scenario (using the load_test)
test = load_test('IOT')

In [3]:

# In order to implement a scenario with respect to the baseline, we can clone the baseline database using the clone function
# and creating a new scenario
test.clone_scenario(
    scenario='baseline', # from which scenario clone  
    name='Final demand increase',# what will be the name of the new scenario
)

In [4]:

#checking the scenarios of the model
print(test.scenarios)

['baseline', 'Final demand increase']

lets increase the local final demand of Italy by 10%

In [5]:

# using the loc function of pd.DataFrames, we can implement a change in the final demand of the new scenario
# we may use the slicer function to do it easily
Y_rows = slicer(matrix='Y',axis= 0,Region=['Italy'],Item=['Agriculture'])
Y_cols = slicer(matrix='Y',axis= 1,Region=['Italy'],Item=['Final Demand'])

In [6]:

Y_new = test.matrices['Final demand increase']['Y']
Y_new.loc[Y_rows,Y_cols]*=1.1

In [7]:

# updating the matrix using the update function to track in the metadata
test.update_scenarios(scenario='Final demand increase',Y=Y_new)

In [8]:

# as the new scenario is cloning all the data from baseline, we need to reset the new scenario to coefficients and re calculte 
# all the flows
test.reset_to_coefficients(scenario='Final demand increase')

In [9]:

# now using the get_data function, we can see the change in the different matrices. For example:
delta_X = test.get_data(
    matrices=['X'],
    scenarios= ['Final demand increase'],
    base_scenario = 'baseline', # requsting the differential results for matrix X between two scenarios,
    format= 'dict',# asking the results in form of a dictionary,
    indeces=False,# not intrested at looking at the units information
    units=False, # not intrested at lookning database information
)

In [10]:

# looking at the changes
print(delta_X)

{'Final demand increase': {'X':                               production
Region Level  Item                      
Italy  Sector Agriculture    2961.338965
              Construction     57.329630
              Manufacturing   597.267115
              Mining            6.332699
              Services        745.379267
              Transport       185.686254
RoW    Sector Agriculture     121.156060
              Construction      5.421589
              Manufacturing   302.857308
              Mining           45.949173
              Services        193.241123
              Transport        36.432992}}

In [11]:

# the output of the get_data function is a comprehensive data on the all the scenarios and matrices asked
print(delta_X['Final demand increase']['X'])

                              production
Region Level  Item                      
Italy  Sector Agriculture    2961.338965
              Construction     57.329630
              Manufacturing   597.267115
              Mining            6.332699
              Services        745.379267
              Transport       185.686254
RoW    Sector Agriculture     121.156060
              Construction      5.421589
              Manufacturing   302.857308
              Mining           45.949173
              Services        193.241123
              Transport        36.432992

In [12]:

# The plot_matrix function can be used to plot the changes of the X
test.plot_matrix(
    matrix='X', # plotting the X matrix
    x='Region_from', # putting the origin regions on the X axis
    color='Sector_from', # colors are defined tby the origin sectors
    base_scenario='baseline', # printing the delta_x with respect to baseline scenario,
    path = 'delta_X.html'
                )