-
Notifications
You must be signed in to change notification settings - Fork 8
/
WEO_data_input.py
159 lines (133 loc) · 4.62 KB
/
WEO_data_input.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
"""
IMF, World Economic Outlook database.
Program reads in the whole thing and arranges it for easy use.
Links
Collection: https://www.imf.org/external/ns/cs.aspx?id=28
April 2015: http://www.imf.org/external/pubs/ft/weo/2015/01/weodata/index.aspx
Prepared for Data Bootcamp course at NYU
* https://github.com/NYUDataBootcamp/Materials
* https://github.com/NYUDataBootcamp/Materials/Code/Lab
Written by Dave Backus, August 2015
Created with Python 3.4
"""
import sys
import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
print('\nPython version: ', sys.version)
print('\nPandas version: ', pd.__version__)
"""
Read data
NB: missing value list is critical, otherwise it doesn't recognize numbers
Also: file is tab-delimited, not xls, despite its name
"""
url1 = 'http://www.imf.org/external/pubs/ft/weo/2015/02/weodata/'
url2 = 'WEOOct2015all.xls'
url = url1 + url2
weo = pd.read_csv(url, sep='\t', thousands=',', na_values=['n/a', '--'])
print('\nWEO database read, dimensions (rows, columns) =', weo.shape)
print('Variable dtypes:\n', weo.dtypes, sep='')
#weo.head().to_csv('weo_head.csv', index=False)
#%%
"""
Create dfs for country codes and variable definitions
"""
country_guide = weo[['ISO', 'Country']].drop_duplicates().set_index('ISO')
variable_guide = weo[['WEO Subject Code',
'Subject Descriptor',
'Subject Notes']
].drop_duplicates().set_index('WEO Subject Code')
variable_guide
#%%
"""
time series plots: one series, multiple countries
"""
variables = ['GGXWDG_NGDP']
countries = ['ARG', 'DEU', 'FRA', 'GRC', 'USA']
sub = weo[weo['ISO'].isin(countries) & weo['WEO Subject Code'].isin(variables)]
some = [3] + list(range(9,44))
sub = sub[some].set_index('Country').T.dropna()
sub
#%%
"""
time series plots: one series, multiple countries
"""
vars = {'NGSD_NGDP': 'Saving', 'NID_NGDP': 'Investment',
'BCA_NGDPD': 'Current Account'}
variables = list(vars)
countries = ['CHN']
sub = weo[weo['ISO'].isin(countries) &
weo['WEO Subject Code'].isin(variables)]
some = [2] + list(range(9,45))
sub = sub[some].set_index('WEO Subject Code').T.dropna().rename(columns=vars)
#sub['zero'] = sub['Saving'] - sub['Investment'] - sub['Current Account']
sub.to_excel('China-Saving-Investment.xls', index=False)
#%%
"""
Cross-section scatterplot
Per capita GDP growth, 2000-15 v 1985-2000
"""
# growth rate data
vlist = ['PPPPC']
xs = weo[weo['WEO Subject Code'].isin(vlist)].set_index('ISO')
# compute 15-year growth rates
g = ['g8500', 'g0015']
xs[g[0]] = 100*(np.log(xs['2000']) - np.log(xs['1985']))/15
xs[g[1]] = 100*(np.log(xs['2015']) - np.log(xs['2000']))/15
# drop unnecessary variables, drop missing values, and sort
xs = xs[g].dropna().sort_index()
# other data: base-year gdp and gdp per capita
base = '2000'
xs['gdppc'] = weo[weo['WEO Subject Code'] == 'PPPPC'].set_index('ISO')[base]
xs['gdp'] = weo[weo['WEO Subject Code'] == 'PPPSH'].set_index('ISO')[base]
# some stats
print('\nStatistics for growth rates')
print('\n', xs.describe(), sep='')
print('\n', xs.corr(), '\n', sep='')
#%%
# scatterplot
plt.scatter(xs[g[0]], xs[g[1]], alpha=0.5)
#plt.scatter(xs[g[0]], xs[g[1]], s=75*xs['gdp'], alpha=0.35)
#plt.scatter(sub1[g[0]], sub1[g[1]], s=100*sub1['gdp']**(1/2), alpha=0.25)
plt.plot(xs[g[0]], xs[g[0]], 'k--', lw=0.5)
plt.xlim(-6, 12)
plt.ylim(-2, 12)
plt.xlabel('Per Capita GDP Growth, 1985-2000')
plt.ylabel('Per Capita GDP Growth, 2000-2015')
plt.show()
#%%
# another one, growth v level
#plt.scatter(sub1[g[0]], sub1[g[1]], alpha=0.5)
plt.scatter(np.log(xs['gdppc'])/np.log(2), xs[g[1]],
s=100*xs['gdp']**(1/2), alpha=0.35)
#plt.scatter(sub1[g[0]], sub1[g[1]], s=100*sub1['gdp']**(1/2), alpha=0.25)
plt.xlim(8, 18)
plt.ylim(-2, 12)
#plt.hlines(0, xmin, xmax, colors='k')
plt.title('GDP growth versus starting point', fontsize=14, loc='left')
plt.xlabel('Per Capita GDP (2000, log2 scale)')
plt.ylabel('Per Capita GDP Growth, 2000-2015')
plt.show()
#%%
# =============================================================================
"""
3. Time series plots
"""
"""
Rearrange so that columns are variables labeled by variable and country code
"""
# select desired countries and variables by their codes
clist = ['USA', 'FRA', 'CHN', 'IND', 'BRA', 'MEX']
vlist = ['PPPPC']
subset = weo[(weo['ISO'].isin(clist)) &
(weo['WEO Subject Code'].isin(vlist))]
#%%
# kill some columns
years = [year for year in range(1980, 2016)]
years_str = [str(year) for year in years]
varlist = ['WEO Subject Code', 'ISO'] + years_str
sub = weo[varlist]
# transpose data
sub = sub.set_index(['WEO Subject Code', 'ISO']).T
sub.index = years
# now do bar chart of GDP per capita