
from IPython.display import Image
Image("figure/img_pollution.jpg")


#!pip install pycountry
#!pip install pycountry_convert


import pandas as pd
import matplotlib.pyplot as plt
import numpy as np
import seaborn as sns
import cufflinks as cf
import geopandas as gpd
import plotly.express as px
import plotly.graph_objects as go

from pycountry import countries
import pycountry_convert as pc
import warnings

warnings.filterwarnings("ignore")

%matplotlib inline

from plotly.offline import download_plotlyjs,init_notebook_mode, plot, iplot
init_notebook_mode(connected=True)
cf.go_offline()


df_death_world  =pd.read_csv("data/number-of-deaths-by-risk-factor.csv",sep=";")
df_death_world= df_death_world.drop(['Unnamed: 38','Unnamed: 39'],axis=1)
df_death_world.head()


head_col = ["Entity","Code","Year"]

colonne_d = []
for col in df_death_world.columns[3:]:
    col_name = col.split("-")
    if len(col_name)>1:
        colonne_d.append(col_name[1])
    else:
        colonne_d.append(col_name)

col_final = head_col + colonne_d
col_final[-1] = "Outdoor air pollution"
colonne_d = col_final[3:]


df_death_world.columns = col_final
#df_death_world.head()


colonne_d = np.asarray(colonne_d).flatten().tolist()
df_death = df_death_world.groupby(["Year"])[colonne_d].sum()
df_death = df_death.iloc[2:,:]

df_death.head()


test_2015 = df_death[df_death.index=="2015"]
test_2015 = ((test_2015.T).sort_values(by="2015",ascending=False)).iloc[:20,:]



fig = px.bar(test_2015,x="2015",y=test_2015.index,color=test_2015.index,text="2015",
             labels={"index":" ","color":" Cause de la mortalité","2015":"Année 2015"},
            color_continuous_scale=px.colors.sequential.Inferno,
             color_continuous_midpoint=2)

fig.update_traces(texttemplate='%{text:.3s}',textposition="outside")
fig.update_layout(uniformtext_minsize=8)


test_1990 = df_death[df_death.index=="1990"]
test_1990 = ((test_1990.T).sort_values(by="1990",ascending=False)).iloc[:20,:]



fig = px.bar(test_1990,x="1990",y=test_1990.index,color=test_1990.index,text="1990",labels={"index":" ","1990":" Année 1990"}
            )
fig.update_traces(texttemplate='%{text:.3s}',textposition="outside")
fig.update_layout(uniformtext_minsize=8)


plt.figure(figsize=(15,5))
plt.plot(df_death[' Air pollution '],label="nombre de Décès")
plt.legend()
plt.xticks(rotation=90)
plt.yticks([19.5e6,20e6,20.5e6,21e6],["19.5M","20M","20.5M","21M"])
plt.ylabel("Nombre en million")
plt.title("Evolution du nombre de décès causé par la pollution de l'air dans le monde")
plt.show()


df_death_ = df_death_world[['Entity', 'Code', 'Year',"Outdoor air pollution"]].rename(columns={"Outdoor air pollution":"death_count"})

for col in df_death_.select_dtypes("object"):
    df_death_[col]= df_death_[col].astype("string")

    
df_death_ = df_death_.loc[(df_death_["Year"]>='1990') ]

df_death_ = df_death_.loc[(df_death_["Year"]<='2016') ]

#df_death_



df_CO2_coutry = pd.read_csv("data/annual-co2-emissions-per-country.csv")

for col in df_CO2_coutry.select_dtypes("object"):
    df_CO2_coutry[col]= df_CO2_coutry[col].astype("string")

df_CO2_coutry["Year"] = df_CO2_coutry.Year.astype("string") 

df_CO2_coutry = df_CO2_coutry.loc[(df_CO2_coutry["Year"]>='1990') ]

df_CO2_coutry = df_CO2_coutry.loc[(df_CO2_coutry["Year"]<='2016') ]
df_CO2_coutry=(df_CO2_coutry.reset_index()).drop("index",axis=1)
#df_CO2_coutry


df_1 = pd.merge(df_death_,df_CO2_coutry, on=['Entity', 'Code', 'Year'])
#df_1





df_CH4_coutry = pd.read_csv("data/methane-emissions.csv")

for col in df_CH4_coutry.select_dtypes("object"):
    df_CH4_coutry[col]= df_CH4_coutry[col].astype("string")

df_CH4_coutry["Year"] = df_CH4_coutry.Year.astype("string") 


df_CH4_coutry=((df_CH4_coutry.reset_index()).drop("index",axis=1)).rename(columns={'Total including LUCF (CH4 emissions, CAIT)':'CH4'})
df_CH4_coutry = df_CH4_coutry.loc[(df_CH4_coutry["Year"]<='2016') ]
#df_CH4_coutry

df_2 = pd.merge(df_1,df_CH4_coutry, on=['Entity', 'Code', 'Year'])

#df_2



df_N20_coutry = pd.read_csv("data/nitrous-oxide-emissions.csv")

for col in df_N20_coutry.select_dtypes("object"):
    df_N20_coutry[col]= df_N20_coutry[col].astype("string")

df_N20_coutry["Year"] = df_N20_coutry.Year.astype("string") 


df_N20_coutry=((df_N20_coutry.reset_index()).drop("index",axis=1)).rename(columns={'Total including LUCF (N2O emissions, CAIT)':'N2O'})
#df_N20_coutry


df_3 = pd.merge(df_2,df_N20_coutry, on=['Entity', 'Code', 'Year'])
#df_3





df_O3_coutry = pd.read_csv("data/ozone-o3-concentration-in-ppb.csv")

for col in df_O3_coutry.select_dtypes("object"):
    df_O3_coutry[col]= df_O3_coutry[col].astype("string")

df_O3_coutry["Year"] = df_O3_coutry.Year.astype("string") 


df_O3_coutry=((df_O3_coutry.reset_index()).drop("index",axis=1)).rename(columns={'Ozone concentration-StateofGlobalAir':'O3'})
#df_O3_coutry

df_4 = pd.merge(df_3,df_O3_coutry, on=['Entity', 'Code', 'Year'])
#df_4





df_PM25_coutry = pd.read_csv("data/PM25-air-pollution.csv")

for col in df_PM25_coutry.select_dtypes("object"):
    df_PM25_coutry[col]= df_PM25_coutry[col].astype("string")

df_PM25_coutry["Year"] = df_PM25_coutry.Year.astype("string") 


df_PM25_coutry=((df_PM25_coutry.reset_index()).drop("index",axis=1)).rename(columns={'PM2.5 air pollution, mean annual exposure (micrograms per cubic meter)':'PM25'})
#df_PM25_coutry

df_5 =pd.merge(df_4,df_PM25_coutry, on=['Entity', 'Code', 'Year'])
df_5["Year"] = df_5["Year"].astype("int")
df_5.head()


fig = px.choropleth(df_5,
                   locationmode = "country names",
                   locations =df_5["Entity"],
                   color = df_5["death_count"] ,
                    labels={"death_count":"Nombre de décès"},
                   animation_frame="Year"
                    )
fig.update_layout(title="Pourcentage de décès par pays",template="plotly_dark")
fig.show()


fig = px.choropleth(df_5,
                   locationmode = "country names",
                   locations =df_5["Entity"],
                   color = df_5["N2O"] ,
                   animation_frame="Year")
fig.update_layout(title="Concentration de N2O par pays")
fig.show()


fig = px.choropleth(df_5,
                   locationmode = "country names",
                   locations =df_5["Entity"],
                   color = df_5["CH4"] ,
                   animation_frame="Year")
fig.update_layout(title="Concentration du méthane(CH4)")
fig.show()


fig = px.choropleth(df_5,
                   locationmode = "country names",
                   locations =df_5["Entity"],
                   color = df_5["Annual CO2 emissions"] ,
                   animation_frame="Year")
fig.update_layout(title="Concentration en CO2 dans l'atmosphère")
fig.show()


fig = px.choropleth(df_5,
                   locationmode = "country names",
                   locations =df_5["Entity"],
                   color = df_5["O3"] ,
                   animation_frame="Year")
fig.update_layout(title="Concentration d'ozone(O3)",template="plotly_dark")
fig.show()


fig = px.choropleth(df_5,
                   locationmode = "country names",
                   locations =df_5["Entity"],
                   color = df_5["PM25"] ,
                   animation_frame="Year")
fig.update_layout(title="Concentration des particules fines (PM2.5)",template="plotly_dark")
fig.show()


from sklearn.preprocessing import StandardScaler


#recuperation de données et normalisation des données

df_acp = df_5.groupby("Entity")[['death_count', 'Annual CO2 emissions', 'CH4',
       'N2O', 'O3', 'PM25']].mean()



#NORMALISATION
scaler = StandardScaler()
df_acp_norm = scaler.fit_transform(df_acp)

df_acp_norm = pd.DataFrame(df_acp_norm, columns=['death_count', 'Annual CO2 emissions', 'CH4',
       'N2O', 'O3', 'PM25'],index=df_acp.index)
df_acp_norm = df_acp_norm.reset_index()
df_acp_norm


df_acp = df_acp.reset_index()

f = px.choropleth(df_acp,
              locationmode ="country names",
              locations=df_acp["Entity"],
              color=df_acp["death_count"],
              labels={"death_count":"nombre de décès"}
              
)
f.update_layout(title="Nombre de décès moyen de 1990 à 2015")
f.show()


df_acp_norm.to_csv("data/df_acp_norm.csv")


Image("figure/inertie.png",height=500,width=600)


Image("figure/individu.png")


Image("figure/variables.png")


Image("figure/N2O_corr.png",width=500,height=300)


Image("figure/CO2_corr.png",width=500,height=300)


Image("figure/CH4_corr.png",width=500,height=300)


Image("figure/foret.jpg",width=900)


import pandas as pd
import matplotlib.pyplot as plt


df_ph = pd.read_csv("data/NTN-All-cy.csv",sep=";")


df_ph = df_ph[[ 'yr', 'Ca',
       'Mg', 'K', 'Na', 'NH4', 'NO3', 'Cl', 'SO4', 'ppt','pH']]

plt.figure(figsize=(15,5))
df_ph[['yr','pH']].groupby("yr")["pH"].mean().plot(legend=True,xticks=range(1978,2020),rot=90)
plt.axhline(5.6,color="red",linestyle='--',label="valeur référence")
plt.legend()
plt.xlabel("Année")
plt.ylabel("Valeur")
#plt.fill_between(range(1978,2020),y1=0,y2=df_ph.pH,color="cyan")
plt.ylim(4.4,5.7)
plt.xlim(1978,2019)
plt.show()


# Pour faire une classification: O si pH<5 et 1 sinon 

df_ph_brute = df_ph[[ 'Ca',
       'Mg', 'K', 'Na', 'NH4', 'NO3', 'Cl', 'SO4', 'ppt','pH']]
df_ph_brute["acidité"] = df_ph_brute["pH"].apply(lambda x: 0 if x<5 else 1)
df_ph_brute = df_ph_brute.drop(["pH"],axis=1)


df_ph = df_ph.groupby("yr")[['Ca','Mg', 'K', 'Na', 'NH4', 'NO3', 'Cl', 'SO4', 'ppt','pH']].mean()
df_ph.head()


df_ph.to_csv("data/data_pH.csv")


plt.figure(figsize=(15,5))
df_ph.corr()["pH"].sort_values(ascending=True).plot.barh(colormap="turbo_r")
plt.legend()
plt.xlabel('valeur')
plt.title("La corrélation de chaque variable avec la variable pH")
plt.show()


Image("figure/ph.png",width=700,height=700)


df_ph[['Ca','Mg', 'K', 'Na', 'NH4', 'NO3', 'Cl', 'SO4']].plot(figsize=(15,6))

plt.xlabel("Année")
plt.xticks(range(1978,2020),rotation=90)
plt.show()


Image("figure/NO3_corr.png",width=500,height=300)


Image("figure/SO4_corr.png",width=500,height=300)


import pandas as pd
import matplotlib.pyplot as plt
import numpy as np

df_ph = pd.read_csv("data/NTN-All-cy.csv",sep=";")


df_ph = df_ph.groupby("yr")[['Ca','Mg', 'K', 'Na', 'NH4', 'NO3', 'Cl', 'SO4', 'ppt','pH']].mean()


plt.figure(figsize=(15,5))
df_ph.drop(['ppt'],axis=1).corr()["pH"].sort_values(ascending=True).plot.barh(colormap="turbo_r")
plt.legend()
plt.xlabel('valeur')
plt.title("La corrélation de chaque variable avec la variable pH")
plt.show()


target = df_ph['pH'].values
data_ph = df_ph.drop(['ppt','pH'],axis=1)
data_ph.head()


from sklearn.preprocessing import StandardScaler
from sklearn.model_selection import train_test_split,KFold,cross_val_score
from sklearn.metrics import accuracy_score
from sklearn.linear_model import LinearRegression
from sklearn.svm import SVR


X_train, X_test, y_train, y_test = train_test_split(data_ph,target,test_size=0.2,random_state=0)


model1 = SVR(C=100)

cv =KFold(5)
cross_val_score(model1, X_train,y_train,cv=cv)

array([0.90670375, 0.78967753, 0.89871238, 0.31067534, 0.89329718])


from sklearn.model_selection import GridSearchCV


param = {"C":[20,30,40,50],"degree":[3,4,5]}

model_svr = SVR()
grid = GridSearchCV(model_svr,param_grid=param,cv=5)


grid.fit(X_train,y_train)

grid.best_score_


model_svr = grid.best_estimator_
model_svr


import pandas as pd
import matplotlib.pyplot as plt
import numpy as np

# Ignorer les warnings
import warnings
warnings.filterwarnings("ignore")


df_00 = pd.read_csv("GlobalLandTemperatures_GlobalLandTemperaturesByCountry.csv")
df_00 = df_00.dropna()

#changer les types de dennées
df_00["dt"] = pd.to_datetime(df_00["dt"])
df_00["Country"] = df_00["Country"].astype("string")


#resupération de l'année
df_00["year"] = df_00["dt"].dt.year
df_00 = (df_00.groupby(["year"])[['AverageTemperature']].mean()).rename(columns={'AverageTemperature':"Temperature moyenne"})
df_00.plot(figsize=(13,5),grid=True)
plt.xlabel("Année")
plt.ylabel("valeur en °C")
plt.title("Evolution de la température moyenne mondiale")
plt.show()


df_00.to_csv("data/ts_R.csv")


from statsmodels.tsa.stattools import adfuller,kpss


def kpss_test(timeseries):
    print ('Results of KPSS Test:')
    kpsstest = kpss(timeseries, regression='c', nlags="auto")
    kpss_output = pd.Series(kpsstest[0:3], index=['Test Statistic','p-value','Lags Used'])
    for key,value in kpsstest[3].items():
        kpss_output['Critical Value (%s)'%key] = value
    print (kpss_output)


kpss_test(df_00)

Results of KPSS Test:
Test Statistic            2.291169
p-value                   0.010000
Lags Used                10.000000
Critical Value (10%)      0.347000
Critical Value (5%)       0.463000
Critical Value (2.5%)     0.574000
Critical Value (1%)       0.739000
dtype: float64

/Users/christianotheddynasa/opt/anaconda3/lib/python3.8/site-packages/statsmodels/tsa/stattools.py:1881: InterpolationWarning: The test statistic is outside of the range of p-values available in the
look-up table. The actual p-value is smaller than the p-value returned.

  warnings.warn(


df_PH_diff = df_00.diff().dropna(axis=0)

kpss_test(df_PH_diff)

Results of KPSS Test:
Test Statistic           0.162877
p-value                  0.100000
Lags Used                8.000000
Critical Value (10%)     0.347000
Critical Value (5%)      0.463000
Critical Value (2.5%)    0.574000
Critical Value (1%)      0.739000
dtype: float64

/Users/christianotheddynasa/opt/anaconda3/lib/python3.8/site-packages/statsmodels/tsa/stattools.py:1885: InterpolationWarning: The test statistic is outside of the range of p-values available in the
look-up table. The actual p-value is greater than the p-value returned.

  warnings.warn(


df_PH_diff.plot(figsize=(12,5),legend=True,grid=True)
plt.title("La série stationnaire après une différenciation")
plt.show()


def adf_test(timeseries):
    print ('Results of Dickey-Fuller Test:')
    dftest = adfuller(timeseries, autolag='AIC')
    dfoutput = pd.Series(dftest[0:4], index=['Test Statistic','p-value','#Lags Used','Number of Observations Used'])
    for key,value in dftest[4].items():
       dfoutput['Critical Value (%s)'%key] = value
    print (dfoutput)


adf_test(df_PH_diff)

Results of Dickey-Fuller Test:
Test Statistic                -7.766162e+00
p-value                        9.176211e-12
#Lags Used                     5.000000e+00
Number of Observations Used    2.600000e+02
Critical Value (1%)           -3.455754e+00
Critical Value (5%)           -2.872721e+00
Critical Value (10%)          -2.572728e+00
dtype: float64


from statsmodels.graphics.tsaplots import plot_acf,plot_pacf


ax1 = plot_acf(df_PH_diff)


ax2 = plot_pacf(df_PH_diff)


Image("figure/prevision.png")

	Entity	Code	Year	Deaths - Diet low in vegetables - Sex: Both - Age: All Ages (Number)	Deaths - Diet low in whole grains - Sex: Both - Age: All Ages (Number)	Deaths - Diet low in nuts and seeds - Sex: Both - Age: All Ages (Number)	Deaths - Diet low in calcium - Sex: Both - Age: All Ages (Number)	Deaths - Unsafe sex - Sex: Both - Age: All Ages (Number)	Deaths - No access to handwashing facility - Sex: Both - Age: All Ages (Number)	Deaths - Child wasting - Sex: Both - Age: All Ages (Number)	...	Deaths - High body-mass index - Sex: Both - Age: All Ages (Number)	Deaths - Low bone mineral density - Sex: Both - Age: All Ages (Number)	Deaths - Diet low in fruits - Sex: Both - Age: All Ages (Number)	Deaths - Diet low in legumes - Sex: Both - Age: All Ages (Number)	Deaths - Low birth weight for gestation - Sex: Both - Age: All Ages (Number)	Deaths - Unsafe water source - Sex: Both - Age: All Ages (Number)	Deaths - Unsafe sanitation - Sex: Both - Age: All Ages (Number)	Deaths - Household air pollution from solid fuels - Sex: Both - Age: All Ages (Number)	Deaths - Air pollution - Sex: Both - Age: All Ages (Number)	Deaths – Outdoor air pollution (all ages) (IHME)
0	Afghanistan	AFG	1990	7678.717644	11381.377345	7299.867330	181.287701	387.167582	5412.314513	22778.849249	...	7701.581280	374.844056	8538.964137	3418.965194	12168.564631	7554.049543	5887.747628	22388.497233	26598.006727	4383.83
1	Afghanistan	AFG	1991	7789.773033	11487.832390	7386.764303	184.251584	394.448285	5287.891103	22292.691113	...	7747.774903	379.854237	8642.847151	3457.398250	12360.635374	7359.676749	5732.770160	22128.758206	26379.532218	4426.36
2	Afghanistan	AFG	1992	8083.234634	11866.235572	7640.628526	191.318450	422.453302	5506.657363	23102.197940	...	7991.018971	388.130434	8961.526496	3586.490291	13459.593716	7650.437822	5954.804987	22873.768789	27263.127914	4568.91
3	Afghanistan	AFG	1993	8452.242405	12335.961682	7968.311853	199.767293	448.328317	7104.620351	27902.669960	...	8281.564408	405.577931	9377.118485	3746.266658	18458.429133	10270.731383	7986.736613	25599.756284	30495.561499	5080.29
4	Afghanistan	AFG	1994	8755.372045	12672.950191	8244.368430	206.619714	464.506355	8051.515953	32929.005932	...	8472.297780	415.349195	9688.448554	3869.288402	19958.388545	11409.177112	8863.010065	28013.167200	33323.161401	5499.23

	Diet low in vegetables	Diet low in whole grains	Diet low in nuts and seeds	Diet low in calcium	Unsafe sex	No access to handwashing facility	Child wasting	Child stunting	Diet high in red meat	Diet low in fiber	...	High body	Low bone mineral density	Diet low in fruits	Diet low in legumes	Low birth weight for gestation	Unsafe water source	Unsafe sanitation	Household air pollution from solid fuels	Air pollution	Outdoor air pollution
Year
1990	5.579176e+06	9.327172e+06	6.197559e+06	464908.837283	2.190842e+06	5.446405e+06	1.475378e+07	4.376233e+06	62371.984805	2.615823e+06	...	9.661706e+06	667884.089373	8.127860e+06	1.770209e+06	8.452891e+06	9.163278e+06	7.095039e+06	1.138135e+07	1.989655e+07	9160523.23
1991	5.624173e+06	9.460549e+06	6.273026e+06	476388.796321	2.549163e+06	5.416547e+06	1.448716e+07	4.275670e+06	63262.304620	2.649371e+06	...	9.933452e+06	682154.369880	8.230030e+06	1.792674e+06	8.389022e+06	9.108674e+06	7.044254e+06	1.125514e+07	2.000852e+07	9408004.53
1992	5.709048e+06	9.654211e+06	6.384863e+06	488270.933365	2.963181e+06	5.331182e+06	1.411180e+07	4.156469e+06	63954.940218	2.699332e+06	...	1.027215e+07	697621.411744	8.384543e+06	1.829575e+06	8.318822e+06	8.933712e+06	6.900140e+06	1.114029e+07	2.011810e+07	9638559.84
1993	5.852533e+06	9.975794e+06	6.588321e+06	503228.195622	3.407305e+06	5.246986e+06	1.376073e+07	4.046425e+06	65504.189105	2.783902e+06	...	1.080256e+07	716722.074874	8.628856e+06	1.909508e+06	8.223404e+06	8.770290e+06	6.761211e+06	1.096505e+07	2.022139e+07	9919298.89
1994	5.932201e+06	1.018312e+07	6.700211e+06	513642.030130	3.879726e+06	5.165670e+06	1.345567e+07	3.946716e+06	66156.677107	2.835616e+06	...	1.120452e+07	732026.486869	8.802091e+06	1.955549e+06	8.084732e+06	8.613781e+06	6.627828e+06	1.080272e+07	2.024440e+07	10106396.81

	Entity	Code	Year	death_count	Annual CO2 emissions	CH4	N2O	O3	PM25
0	Afghanistan	AFG	1990	4383.83	2.603238	8.97	3.25	62	60.936690
1	Afghanistan	AFG	1995	5641.29	1.230463	9.15	3.07	62	56.136628
2	Afghanistan	AFG	2000	5587.68	0.757672	10.59	3.62	62	56.092028
3	Afghanistan	AFG	2005	6189.71	1.303267	11.68	3.97	62	43.234032
4	Afghanistan	AFG	2010	6154.12	8.397779	30.04	6.03	63	46.536809

	Entity	death_count	Annual CO2 emissions	CH4	N2O	O3	PM25
0	Afghanistan	-0.123658	-0.233731	-0.111301	-0.228329	0.589413	1.025517
1	Albania	-0.192597	-0.234177	-0.324418	-0.318999	1.119308	-0.714101
2	Algeria	-0.067475	-0.070569	-0.019187	-0.164475	1.086865	0.042081
3	Andorra	-0.201928	-0.239771	-0.353320	-0.348636	0.600227	-0.886610
4	Angola	-0.161977	-0.209349	0.060832	0.123168	2.189912	0.206867
...	...	...	...	...	...	...	...
175	Venezuela	-0.114305	0.005135	0.282677	-0.049444	-0.838059	-0.127359
176	Vietnam	0.058450	-0.094586	0.250729	0.106208	-0.308164	-0.162208
177	Yemen	-0.139674	-0.214969	-0.300938	-0.282857	-0.178394	1.090758
178	Zambia	-0.181725	-0.236269	0.548885	0.336800	0.459643	-0.061778
179	Zimbabwe	-0.178517	-0.221289	-0.261513	-0.200716	0.005447	-0.284863

	Ca	Mg	K	Na	NH4	NO3	Cl	SO4	ppt	pH
yr
1978	0.171714	0.056476	0.041524	0.445286	0.224571	1.191286	0.456286	2.130143	26.932905	4.608095
1979	0.212263	0.045237	0.037711	0.405474	0.257053	1.283289	0.324553	2.312816	86.041553	4.671368
1980	0.257988	0.062000	0.055358	0.413753	0.299160	1.372395	0.590247	2.131568	65.082654	4.755185
1981	0.152770	-0.017410	-0.048300	0.151630	0.173390	1.025720	0.291960	1.920480	79.181920	4.639180
1982	0.209575	0.053920	0.039991	0.210575	0.233389	1.019973	0.368009	1.656947	93.613115	4.826690

LA POLLUTION DE L'AIR¶

I- Les pertes humaines causées par la pollution de l'air dans le monde¶

1- Evolution de nombre de décès dans le monde¶

- Nombre de décès en 2015 dans le monde:¶

- Nombre de décès en 1990 dans le monde:¶

2-Analyse des facteurs de décès¶

Analyse en composante principale:¶

a-Visualisation générale des variables¶

b- Application de ACP normée¶

Description du resultat de l'ACP¶

a) Distribution de l'inertie¶

Test de la nullité de corrélation¶

II-Acidité de la pluie et ses effets sur l'ecosystème¶

Test de la nullité de corrélation¶

III-Prédiction de l'acidité de la précipitation¶

MODELISATION:¶

Corrélation des données avec la variable à expliquer pH¶

Par validation croisée:¶

Calibrage des hyperparamètres du modèle Support Vector Machine¶

IV- Prévision de la température mondiale¶

Stationnarité de la série¶

a) Test de KPSS¶

b) Test de Dickey-Fuller Augmenté¶

Proposition d'un modèle¶

Autocorrélation et autocorrélation partielle¶

V- Conclusion:¶

VI- Bibliographie:¶