def create_table(models, path):
path = Path(path)
stargazer = Stargazer(models)
if path.suffix == ".tex":
table = stargazer.render_latex()
elif path.suffix == ".html":
table = stargazer.render_html()
else:
raise NotImplementedError
with open(path, "w") as file:
file.write(table)
def table3_7(df, regression_type):
df_3_7E = df[[
'ethnicity_C2', 'ethnicity_instrument_C2_thresh', 'ethnicity_I',
'lnpopulation', 'lnGDP_pc', 'protestants', 'muslims', 'catholics',
'latitude', 'LOEnglish', 'LOGerman', 'LOSocialist', 'lnArea',
'LOScandin', 'democ', 'mtnall', 'RulLaw'
]].dropna(axis=0)
df_3_7L = df[[
'language_C2', 'language_instrument_C2_thresh', 'language_I',
'lnpopulation', 'lnGDP_pc', 'protestants', 'muslims', 'catholics',
'latitude', 'LOEnglish', 'LOGerman', 'LOSocialist', 'lnArea',
'LOScandin', 'democ', 'mtnall', 'RulLaw'
]].dropna(axis=0)
df_3_7R = df[[
'religion_C2', 'religion_instrument_C2_thresh', 'religion_I',
'lnpopulation', 'lnGDP_pc', 'protestants', 'muslims', 'catholics',
'latitude', 'LOEnglish', 'LOGerman', 'LOSocialist', 'lnArea',
'LOScandin', 'democ', 'mtnall', 'RulLaw'
]].dropna(axis=0)
exo = sm.add_constant(df_3_7E[[
'ethnicity_C2', 'ethnicity_I', 'lnpopulation', 'lnGDP_pc',
'protestants', 'muslims', 'catholics', 'latitude', 'LOEnglish',
'LOGerman', 'LOSocialist', 'LOScandin', 'lnArea', 'democ', 'mtnall'
]])
exo2 = sm.add_constant(df_3_7E[['ethnicity_C2', 'ethnicity_I']])
exo3 = sm.add_constant(df_3_7L[[
'language_C2', 'language_I', 'lnpopulation', 'lnGDP_pc', 'protestants',
'lnArea', 'muslims', 'catholics', 'latitude', 'LOEnglish', 'LOGerman',
'LOSocialist', 'LOScandin', 'democ', 'mtnall'
]])
exo4 = sm.add_constant(df_3_7L[['language_C2', 'language_I']])
exo5 = sm.add_constant(df_3_7R[[
'religion_C2', 'religion_I', 'lnpopulation', 'lnGDP_pc', 'protestants',
'muslims', 'catholics', 'latitude', 'LOEnglish', 'LOGerman',
'LOSocialist', 'lnArea', 'democ', 'mtnall'
]])
exo6 = sm.add_constant(df_3_7R[['religion_C2', 'religion_I']])
if regression_type == 'IV2SLS':
reg = IV2SLS(
df_3_7E['RulLaw'], exo,
sm.add_constant(df_3_7E[[
'ethnicity_instrument_C2_thresh', 'ethnicity_I',
'lnpopulation', 'lnGDP_pc', 'protestants', 'muslims',
'catholics', 'latitude', 'LOEnglish', 'LOGerman',
'LOSocialist', 'LOScandin', 'democ', 'mtnall', 'lnArea'
]])).fit()
reg2 = IV2SLS(
df_3_7E['RulLaw'], exo2,
sm.add_constant(
df_3_7E[['ethnicity_instrument_C2_thresh',
'ethnicity_I']])).fit()
reg3 = IV2SLS(
df_3_7L['RulLaw'], exo3,
sm.add_constant(df_3_7L[[
'language_instrument_C2_thresh', 'language_I', 'lnpopulation',
'lnGDP_pc', 'protestants', 'muslims', 'catholics', 'latitude',
'LOEnglish', 'LOGerman', 'LOSocialist', 'LOScandin', 'democ',
'mtnall', 'lnArea'
]])).fit()
reg4 = IV2SLS(
df_3_7L['RulLaw'], exo4,
sm.add_constant(
df_3_7L[['language_instrument_C2_thresh',
'language_I']])).fit()
reg5 = IV2SLS(
df_3_7R['RulLaw'], exo5,
sm.add_constant(df_3_7R[[
'religion_instrument_C2_thresh', 'religion_I', 'lnpopulation',
'lnGDP_pc', 'protestants', 'muslims', 'catholics', 'latitude',
'LOEnglish', 'LOGerman', 'LOSocialist', 'democ', 'mtnall',
'lnArea'
]])).fit()
reg6 = IV2SLS(
df_3_7R['RulLaw'], exo6,
sm.add_constant(
df_3_7R[['religion_instrument_C2_thresh',
'religion_I']])).fit()
elif regression_type == 'OLS':
reg2 = sm.OLS(df_3_7E['RulLaw'], exo2).fit(cov_type='HC1')
reg = sm.OLS(df_3_7E['RulLaw'], exo).fit(cov_type='HC1')
reg4 = sm.OLS(df_3_7L['RulLaw'], exo4).fit(cov_type='HC1')
reg3 = sm.OLS(df_3_7L['RulLaw'], exo3).fit(cov_type='HC1')
reg6 = sm.OLS(df_3_7R['RulLaw'], exo6).fit(cov_type='HC1')
reg5 = sm.OLS(df_3_7R['RulLaw'], exo5).fit(cov_type='HC1')
stargazer = Stargazer([reg2, reg, reg4, reg3, reg6, reg5])
stargazer.covariate_order([
'ethnicity_C2', 'ethnicity_I', 'language_C2', 'language_I',
'religion_C2', 'religion_I', 'lnpopulation', 'lnGDP_pc', 'lnArea',
'protestants', 'muslims', 'catholics', 'latitude', 'LOEnglish',
'LOGerman', 'LOSocialist', 'LOScandin', 'democ', 'mtnall', 'const'
])
stargazer.rename_covariates({
'ethnicity_C2': 'Segregation $\hat{S}$ (ethnicity)',
'ethnicity_I': 'Fractionalization $F$ (ethnicity)',
'language_C2': 'Segregation $\hat{S}$ (language)',
'language_I': 'Fractionalization $F$ (language)',
'religion_C2': 'Segregation $\hat{S}$ (religion)',
'religion_I': 'Fractionalization $F$ (religion)',
'lnpopulation': 'ln (population)',
'lnGDP_pc': 'ln (GDP per capita)',
'lnArea': 'ln (average size of region)',
'protestants': 'Pretestants share',
'muslims': 'Muslmis Share',
'catholics': 'Catholics share',
'latitude': 'Latitude',
'LOEnglish': 'English legal origin',
'LOGerman': 'German legal origin',
'LOSocialist': 'Socialist legal origin',
'LOScandin': 'Scandinavian legal origin',
'democ': 'Democratic tradition',
'mtnall': 'Mountains',
'const': 'Constant'
})
return HTML(stargazer.render_html())
model1a = ols('cnt ~ temp_celsius', data=wbr).fit() #Primero y, luego x aquí
model1a.summary2()
model1b = ols('cnt ~ windspeed_kh', data=wbr).fit()
model1b.summary2(
) #Es significativa también y negativa, pero R2 es mucho menor: solo el 6% depende de la variabilidad en el viento
###
model2 = ols('cnt ~ temp_celsius + windspeed_kh',
data=wbr).fit() #Dos predictores ahora
model2.summary2()
###
wbr.hum.hist() #Describir primero SIEMPRE
model1c = ols('cnt ~ hum', data=wbr).fit()
model1c.summary2()
model3 = ols('cnt ~ temp_celsius + windspeed_kh + hum', data=wbr).fit()
model3.summary2() #Aumenta R2 respecto a model2 y cambian los coeficientes
### Para reportar
#!pip install stargazer
from stargazer.stargazer import Stargazer
stargazer = Stargazer([model1a, model2, model3])
stargazer.render_html()
def main():
#%%
#Load data
os.chdir('/Users/rgreen/Documents/Github/NDVI_Projection/')
oromia = pd.ExcelFile('oromia.xlsx')
arsi = pd.read_excel(oromia, 'arsi')
bale = pd.read_excel(oromia, 'bale')
borena = pd.read_excel(oromia, 'borena')
guji = pd.read_excel(oromia, 'guji')
westarsi = pd.read_excel(oromia, 'westarsi')
arsi.insert(0, 'Time', np.linspace(1,585,585))
bale.insert(0, 'Time', np.linspace(1,585,585))
borena.insert(0, 'Time', np.linspace(1,585,585))
guji.insert(0, 'Time', np.linspace(1,585,585))
westarsi.insert(0, 'Time', np.linspace(1,585,585))
#%%
#dekadal data (P, LST, ET)
ddf = pd.DataFrame()
ddf['D_NDVI'] = arsi.NDVI.diff()[1:]
ddf = ddf.reset_index(drop=True)
ddf['N_P'] = ((max(arsi.NDVI) - (arsi.NDVI[:-1]))* (arsi.P[:-1]))
ddf['N_LST'] = (((arsi.NDVI[:-1]) - min(arsi.NDVI))* (arsi.LST[:-1])).shift(4) #need to shift back lags, use shift not index
ddf['N_ET'] = (((arsi.NDVI[:-1]) - min(arsi.NDVI))* (arsi.ET[:-1])).shift(4)
ddf = pd.DataFrame()
ddf['D_NDVI'] = arsi.NDVI.diff().shift(-1)
ddf = ddf.reset_index(drop=True)
#L_NDVI = arsi.NDVI.shift(-1) #lag
ddf['N_P'] = ((max(arsi.NDVI) - (arsi.NDVI))* (arsi.P))
ddf['N_LST'] = (((arsi.NDVI) - min(arsi.NDVI))* (arsi.LST)).shift(4) #need to shift back lags, use shift not index
ddf['N_ET'] = (((arsi.NDVI) - min(arsi.NDVI))* (arsi.ET)).shift(4)
#shift is (n-1) dekads, ex. shift(4) is 5 dekads lag
#mask
mask_lst = ~np.isnan(ddf.N_LST) & ~np.isnan(ddf.D_NDVI)
mask_et = ~np.isnan(ddf.N_ET) & ~np.isnan(ddf.D_NDVI)
#stats.lingress (X,Y)
slope1, intercept1, r1, p1, std1 = stats.linregress(ddf.N_P, ddf.D_NDVI)
line1 = slope1*ddf.N_P+intercept1
print("r-squared: %f" % r1**2)
slope2, intercept2, r2, p2, std2 = stats.linregress(ddf.N_LST[mask_lst], ddf.D_NDVI[mask_lst])
line2 = slope2*ddf.N_LST+intercept2
print("r-squared: %f" % r2**2)
slope3, intercept3, r3, p3, std3 = stats.linregress(ddf.N_ET[mask_et], ddf.D_NDVI[mask_et])
line3 = slope3*ddf.N_ET+intercept3
print("r-squared: %f" % r3**2)
fig, (ax1, ax2, ax3) = plt.subplots(3, sharey =True)
ax1.scatter(ddf.N_P, ddf.D_NDVI, color = 'darkcyan')
ax1.plot(ddf.N_P, line1, color = 'k')
ax1.set(xlabel = r'$(NDVI_{max} - NDVI_{t-1})*P_{t-1}$', ylabel = '')
ax1.text(16, -0.04, r'y = 0.005x + 0.02', fontsize=8)
ax1.text(16, -0.05, r'$r^2$ = 0.528', fontsize=8)
ax2.scatter(ddf.N_LST, ddf.D_NDVI, color = 'forestgreen')
ax2.set(xlabel = r'$(NDVI_{t-5} - NDVI_{min})*LST_{t-5}$', ylabel = '')
ax2.plot(ddf.N_LST, line2, color = 'k')
ax2.text(9, 0.07, r'y = -0.005x + 0.038', fontsize=8)
ax2.text(9, 0.06, r'$r^2$ = 0.386', fontsize=8)
ax3.scatter(ddf.N_ET, ddf.D_NDVI, color = 'cornflowerblue')
ax3.set(xlabel = r'$(NDVI_{max} - NDVI_{t-5})*ET_{t-5}$', ylabel = '')
ax3.plot(ddf.N_ET, line3, color = 'k')
ax3.text(15, 0.07, r'y = -0.003x + 0.031', fontsize=8)
ax3.text(15, 0.06, r'$r^2$ = 0.414', fontsize=8)
fig.text(0.06, 0.5, r'$\Delta NDVI$', ha='center', va='center', rotation='vertical') #common ylabel
# =============================================================================
#
# sns.set(style="ticks", color_codes=True)
#
# fig = plt.figure()
# sns.regplot(x=ddf.N_P, y=ddf.D_NDVI)
#
# g = sns.PairGrid(ddf, y_vars=["D_NDVI"], x_vars=["N_P", "N_LST", "N_ET"], height=4)
# g.map(sns.regplot, color=".3")
#
# replacements = {'D_NDVI': r'$\Delta NDVI$', 'N_P': '(maxNDVI - NDVIt-1)*Pt-1',
# 'N_LST': '(NDVIt-5 - minNDVI)*LSTt-5', 'N_ET': '(NDVIt-5 - minNDVI)*N_ETt-5'}
#
# for i in range(4):
# for j in range(4):
# xlabel = g.axes[i][j].get_xlabel()
# ylabel = g.axes[i][j].get_ylabel()
# if xlabel in replacements.keys():
# g.axes[i][j].set_xlabel(replacements[xlabel])
# if ylabel in replacements.keys():
# g.axes[i][j].set_ylabel(replacements[ylabel])
#
# =============================================================================
# =============================================================================
#
# r, p = stats.pearsonr(ddf.D_NDVI, ddf.N_P)
# r, p = stats.pearsonr(ddf.D_NDVI, ddf.N_LST)
# r, p = stats.pearsonr(ddf.D_NDVI, ddf.N_LST)
#
# =============================================================================
#%%
X = ddf[['N_P']]
Y = ddf['D_NDVI']
X = sm.add_constant(X)
est = sm.OLS(Y,X, missing = 'drop').fit()
est.summary()
Y_pred = est.predict()
original = Y + arsi.NDVI
predicted = Y_pred + arsi.NDVI[:-2]
result = pd.concat([original, predicted], axis=1)
result.columns= ['Original','Predicted']
rms1 = sqrt(mean_squared_error(result.Original[:-2], result.Predicted[:-2])) #0.0186
X = ddf[['N_P', 'N_LST', 'N_ET']]
Y = ddf['D_NDVI']
X = sm.add_constant(X)
est = sm.OLS(Y,X, missing = 'drop').fit()
est.summary()
Y_pred = est.predict()
original = Y + arsi.NDVI
predicted = Y_pred + arsi.NDVI[:-6]
result = pd.concat([original, predicted], axis=1)
result.columns= ['Original','Predicted']
rms1 = sqrt(mean_squared_error(result.Original[:-6], result.Predicted[:-6])) #0.0186
fig = plt.figure()
plt.plot(arsi.DT, result.Original, 'darkblue', label = 'Original')
plt.plot(arsi.DT, result.Predicted, 'dodgerblue', label = 'Projected')
plt.legend(loc='upper right')
plt.title('Arsi NDVI')
X = ((max(arsi.NDVI) - arsi.NDVI)* arsi.P) #for multivariate regression ddf[['N_P', 'N_LST']]
Y = arsi.NDVI.diff().shift(-1)
X = sm.add_constant(X) #only use when doing first run OLS then remove when fitting prediction
est = sm.OLS(Y,X, missing = 'drop').fit()
est.summary()
Y_pred = est.predict()
Y_pred2 = 0.0047*X - 0.0201
#take delta predictions and convert to forecast
original = Y + arsi.NDVI
original = np.append(result.original, np.nan)
predicted = Y_pred2 + arsi.NDVI
result = pd.concat([original, predicted], axis=1)
result.columns= ['Original','Predicted']
fig = plt.figure()
plt.plot(arsi.DT, result.Original, 'darkblue', label = 'Original')
plt.plot(arsi.DT, result.Predicted, 'dodgerblue', label = 'Projected')
plt.legend(loc='upper right')
plt.title('Arsi NDVI')
rms1 = sqrt(mean_squared_error(result.Original[:-2], result.Predicted[:-2])) #0.0186
#create stargazer model table
stargazer = Stargazer([est])
HTML(stargazer.render_html())
model = sm.OLS(ddf.D_NDVI, sm.add_constant(ddf.N_P)).fit()
#predict values of Y
Y_pred = model.predict()
#summary table
model.summary()