-
Notifications
You must be signed in to change notification settings - Fork 1
/
Copy pathCFS_distribution_generator.py
188 lines (147 loc) · 8.16 KB
/
CFS_distribution_generator.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
#!/usr/bin/env python3
# -*- coding: utf-8 -*-
"""
Created on Thu May 11 12:03:20 2023
@author: xiaodanxu
"""
import pandas as pd
import os, inspect
from sklearn.utils import shuffle
import numpy as np
from statsmodels.stats.weightstats import DescrStatsW
from pandas import read_csv
import matplotlib.pyplot as plt
import seaborn as sns
import warnings
warnings.filterwarnings("ignore")
plt.style.use('ggplot')
sns.set(font_scale=1.2) # larger font
os.chdir('/Users/xiaodanxu/Documents/SynthFirm.nosync/')
# <codecell>
# load 2017 CFS data
CFS_df = pd.read_csv('RawData/CFS/cfs-2017-puf-csv.csv', sep = ',')
print(len(CFS_df))
# load additional input data
naics_lookup = pd.read_csv('RawData/corresp_naics6_n6io_sctg_revised.csv')
# load industry-commodity lookup
sctg_group_lookup = pd.read_csv('RawData/SCTG_Groups_revised.csv')
# load sctg - sctg group lookup
sctg_group_lookup.loc[:, 'SCTG_Code'] = sctg_group_lookup.loc[:, 'SCTG_Code'].astype(str)
sctg_group_lookup.loc[:, 'SCTG_Code'] = sctg_group_lookup.loc[:, 'SCTG_Code'].str.zfill(2)
cfs_to_faf_lookup = pd.read_csv('RawData/CFS_FAF_LOOKUP.csv', sep = ',')
# load CFS zone and FAF zone lookup
# clean CFS data
df_clean = CFS_df.loc[CFS_df.EXPORT_YN == 'N'] # Cleans out international exports
lb_to_ton = 0.0005
df_clean.loc[df_clean['SCTG'] == '15-19', 'SCTG'] = '16'
# value_density_by_sctg_zone.loc[value_density_by_sctg_zone['SCTG'] == '15-19', 'SCTG'] = '16'
to_drop = ['35-38', '25-30', '31-34', '10-14', '06-09',
'20-24', '01-05', '39-43']
df_clean = \
df_clean[~df_clean['SCTG'].isin(to_drop)]
df_clean = pd.merge(df_clean, sctg_group_lookup,
left_on = 'SCTG', right_on = 'SCTG_Code', how = 'left')
df_clean.loc[:, 'SHIPMT_WGHT_scaled'] = df_clean.loc[:, 'SHIPMT_WGHT'] * \
df_clean.loc[:, 'WGT_FACTOR'] * lb_to_ton # in tons
df_clean.loc[:, 'SHIPMT_VALUE_scaled'] = df_clean.loc[:, 'SHIPMT_VALUE'] * df_clean.loc[:, 'WGT_FACTOR']
df_clean.loc[:, 'unitcost'] = df_clean.loc[:, 'SHIPMT_VALUE'] / \
df_clean.loc[:, 'SHIPMT_WGHT'] / lb_to_ton # in $/ton
# <codecell>
# Output 1 - compute value density by SCTG as a surrogate of unit cost
# estimate value density -- no region code
value_density_by_sctg = df_clean.groupby(['SCTG'])[['SHIPMT_WGHT_scaled', 'SHIPMT_VALUE_scaled']].sum()
value_density_by_sctg = value_density_by_sctg.reset_index()
# value_density_by_sctg.head(10)
value_density_by_sctg.loc[:, 'unitcost'] = value_density_by_sctg.loc[:, 'SHIPMT_VALUE_scaled'] / \
value_density_by_sctg.loc[:, 'SHIPMT_WGHT_scaled'] # in $/ton
value_density_by_sctg = value_density_by_sctg[['SCTG', 'unitcost']]
value_density_by_sctg.columns = ['Commodity_SCTG', 'UnitCost']
value_density_by_sctg.to_csv('RawData/CFS/data_unitcost_cfs2017.csv', index = False)
# estimate variation for checking
def weighted_std(df):
stats = DescrStatsW(df['unitcost'], weights=df['SHIPMT_WGHT_scaled'], ddof=1)
w_mean = stats.mean
w_std = stats.std
return(pd.Series([w_mean, w_std]))
value_density_by_sctg_2 = df_clean.groupby(['SCTG']).apply(weighted_std)
value_density_by_sctg_2 = value_density_by_sctg_2.reset_index() # weighted mean and std
#value_density_by_sctg_2.head(10)
# <codecell>
# Output 2 - generate production capacity and unit cost by origin/seller zone
value_density_by_sctg_zone = df_clean.groupby(['SCTG', 'ORIG_CFS_AREA'])[['SHIPMT_WGHT_scaled', 'SHIPMT_VALUE_scaled']].sum()
value_density_by_sctg_zone = value_density_by_sctg_zone.reset_index()
value_density_by_sctg_zone.loc[:, 'unitcost'] = value_density_by_sctg_zone.loc[:, 'SHIPMT_VALUE_scaled'] / \
value_density_by_sctg_zone.loc[:, 'SHIPMT_WGHT_scaled']
value_density_by_sctg_zone = \
value_density_by_sctg_zone[~value_density_by_sctg_zone['SCTG'].isin(to_drop)]
value_density_by_sctg_zone = value_density_by_sctg_zone[['SCTG', 'ORIG_CFS_AREA', 'SHIPMT_WGHT_scaled', 'unitcost']]
value_density_by_sctg_zone.columns = ['Commodity_SCTG', 'ORIG_CFS_AREA', 'Capacity', 'UnitCost']
value_density_by_sctg_zone.head(5)
value_density_by_sctg_zone.to_csv('RawData/CFS/data_unitcost_by_zone_cfs2017.csv', index = False)
# <codecell>
# Output 3 - estimate shipment size (using pre-defined quantile cut-off point)
def weighted_quantile(df, quantile=0.5):
df_sorted = df.sort_values('SHIPMT_WGHT')
cumsum = df_sorted['SHIPMT_WGHT_scaled'].cumsum()
cutoff = df_sorted['SHIPMT_WGHT_scaled'].sum() * quantile
value = df_sorted[cumsum >= cutoff]['SHIPMT_WGHT'].iloc[0]
return(value)
load_by_sctg = df_clean.groupby(['SCTG']).apply(weighted_quantile)
load_by_sctg = load_by_sctg.reset_index()
load_by_sctg.columns = ['Commodity_SCTG', 'SHIPMT_WGHT']
load_by_sctg.to_csv('RawData/CFS/max_load_per_shipment_50percent.csv', index = False)
# <codecell>
# check sample size by sctg
count_by_sctg = df_clean.groupby(['SCTG'])[['SHIPMT_WGHT']].count()
count_by_sctg.columns = ['sample_size']
count_by_sctg = count_by_sctg.reset_index()
print(count_by_sctg)
# <codecell>
# Output 4 - Generate shipment value allocation factor by zone
value_fraction_by_origin = df_clean.groupby(['SCTG', 'ORIG_CFS_AREA'])[['SHIPMT_VALUE_scaled']].sum()
value_fraction_by_origin = value_fraction_by_origin.reset_index()
value_fraction_by_dest = df_clean.groupby(['SCTG', 'DEST_CFS_AREA'])[['SHIPMT_VALUE_scaled']].sum()
value_fraction_by_dest = value_fraction_by_dest.reset_index()
value_fraction_by_origin_faf = pd.merge(value_fraction_by_origin,
cfs_to_faf_lookup,
left_on = 'ORIG_CFS_AREA',
right_on = 'ST_MA', how = 'left')
value_fraction_by_dest_faf = pd.merge(value_fraction_by_dest,
cfs_to_faf_lookup,
left_on = 'DEST_CFS_AREA',
right_on = 'ST_MA', how = 'left')
value_fraction_by_origin_faf = value_fraction_by_origin_faf.dropna(subset = ['FAF'])
value_fraction_by_origin_faf.loc[:, 'value_fraction'] = value_fraction_by_origin_faf.loc[:, 'SHIPMT_VALUE_scaled'] / \
value_fraction_by_origin_faf.groupby('SCTG')['SHIPMT_VALUE_scaled'].transform('sum')
value_fraction_by_dest_faf = value_fraction_by_dest_faf.dropna(subset = ['FAF'])
value_fraction_by_dest_faf.loc[:, 'value_fraction'] = value_fraction_by_dest_faf.loc[:, 'SHIPMT_VALUE_scaled'] / \
value_fraction_by_dest_faf.groupby('SCTG')['SHIPMT_VALUE_scaled'].transform('sum')
output_attr = ['SCTG', 'FAF', 'value_fraction']
output_attr_label = ['Commodity_SCTG', 'FAF', 'value_fraction']
value_fraction_by_origin_faf = value_fraction_by_origin_faf[output_attr]
value_fraction_by_origin_faf.columns = output_attr_label
value_fraction_by_dest_faf = value_fraction_by_dest_faf[output_attr]
value_fraction_by_dest_faf.columns = output_attr_label
value_fraction_by_origin_faf.to_csv('RawData/CFS/producer_value_fraction_by_faf.csv', index = False)
value_fraction_by_dest_faf.to_csv('RawData/CFS/consumer_value_fraction_by_faf.csv', index = False)
# <codecell>
# Output 5 - seasonal allocation factor for national freight trip generation -- future use
shipment_by_season = df_clean.groupby(['SCTG_Group', 'SCTG_Name', 'QUARTER'])[['SHIPMT_WGHT_scaled', 'WGT_FACTOR']].sum()
shipment_by_season = shipment_by_season.reset_index()
shipment_by_season.loc[:, 'weight_fraction'] = \
shipment_by_season.loc[:, 'SHIPMT_WGHT_scaled'] / \
shipment_by_season.groupby('SCTG_Group')['SHIPMT_WGHT_scaled'].transform('sum')
shipment_by_season.loc[:, 'count_fraction'] = \
shipment_by_season.loc[:, 'WGT_FACTOR'] / \
shipment_by_season.groupby('SCTG_Group')['WGT_FACTOR'].transform('sum')
shipment_by_season.loc[:, 'SCTG_Group'] = shipment_by_season.loc[:, 'SCTG_Group'].astype(int)
shipment_by_season_out = shipment_by_season[['SCTG_Group', 'SCTG_Name', 'QUARTER', 'weight_fraction']]
shipment_by_season_out.head(5)
shipment_by_season_out.to_csv('RawData/CFS/seasonal_allocation_factor_by_sctg.csv')
sns.barplot(data=shipment_by_season_out,
x="SCTG_Name", y="weight_fraction", hue="QUARTER")
plt.legend(loc='upper right', bbox_to_anchor=(1.2,1.02),
title = 'Quarter')
plt.xlabel('SCTG Group')
plt.ylabel('Fraction of shipment by weight')
plt.savefig('RawData/CFS/CFS2017_seasonal_variation.png', dpi = 200, bbox_inches = 'tight')