NO2

SO2

PM2.5

Monthly mean values (bands for annual mean and maxima) for three contaminants (μg/m3)

Monthly mean values (bands for annual mean and maxima) for three contaminants (μg/m3)

Monthly mean values (bands for annual mean and maxima) for NO2 (μg/m3)

Monthly mean values (bands for annual mean and maxima) for PM2.5 (μg/m3)

Monthly mean values (bands for annual mean and maxima) for SO2 (μg/m3)

Relative Humidity (%): Monthly and yearly (mean and median, respectively) values

Monthly Precipitation (mm), with yearly maximum tendency over time

Monthly Precipitation (mm), with yearly maximum tendency over time

Temperature (°C, minimum-mean-maximum)

Relative Humidity (%): Monthly and yearly (mean and median, respectively) values

Precipitation (mm)

Temp [minimum-mean-maximum]

J:00; Yearly cases (mean and median) for Guatemala Department

J:00; Yearly cases for Guatemala Department

J:02; Yearly cases (mean and median) for Guatemala Department

J:02; Yearly cases for Guatemala Department

J:03; Yearly cases (mean and median) for Guatemala Department

J:03; Yearly cases for Guatemala Department

J:06; Yearly cases (mean and median) for Guatemala Department

J:06; Yearly cases for Guatemala Department

Plot

Plot

J:00 Ciudad Guatemala

J:03 Ciudad Guatemala

J:20 Ciudad Guatemala

J:02 Ciudad Guatemala

J:06 Ciudad Guatemala

Cases of Respiratory Diseases per month (2012-2019)

Contrastes de Cases of enfermedades respiratorias

Humedad Relativa

Contrastes de Cases of enfermedades respiratorias

Precipitación Pluvial

Contrastes de Cases of enfermedades respiratorias

Temp [min, media, max]

Contrastes de Cases of enfermedades respiratorias

Contaminantes (Annual)

Enfermedades: Casos acumulados

Sexo Femenino

Enfermedades: Casos acumulados por código

Sexo masculino

Municipio de Guatemala Contrastado

Plot

Plot

Plot

Plot

Plot

Plot

Plot

J:01 Ciudad Guatemala

J:04 Ciudad Guatemala

J:21 Ciudad Guatemala

J:22 Ciudad Guatemala

Cases of Acute Respiratory Diseases per month (2012-2019)

Contrastes de Cases of enfermedades respiratorias

Contaminantes (Annual)

Enfermedades: Casos acumulados

Sexo Femenino

Enfermedades: Casos acumulados por código

Sexo masculino

Municipio de Guatemala Contrastado

Plot

Plot

Plot

Plot

Plot

Plot

Plot

J:30 Ciudad Guatemala

J:32 Ciudad Guatemala

J:44 Ciudad Guatemala

J:45 Ciudad Guatemala

Cases of Chronic Respiratory Diseases per month (2012-2019)

Contrastes de Cases of enfermedades respiratorias

Contaminantes (Annual)

Enfermedades: Casos acumulados

Sexo Femenino

Enfermedades: Casos acumulados por código

Sexo masculino

Municipio de Guatemala Contrastado

Plot

Plot

Plot

Plot

Plot

J:18 Ciudad Guatemala

J:40 Ciudad Guatemala

J:98 Ciudad Guatemala

Cases of Not Specified Respiratory Diseases per month (2012-2019)

Contrastes de Cases of enfermedades respiratorias

Contaminantes (Annual)

Enfermedades: Casos acumulados

Sexo Femenino

Enfermedades: Casos acumulados por código

Sexo masculino

Personas afectadas

Contrastes de Cases of enfermedades respiratorias

Personas afectadas

Contrastes de Cases of enfermedades respiratorias Agudas

Personas afectadas

Contrastes de Cases of enfermedades respiratorias

Componentes principales

***VECTORS

                   PC1      PC2     r2 Pr(>r)    
Gatherings    -0.80006  0.59992 0.6989  0.001 ***
Fires         -0.28550 -0.95838 0.1422  0.003 ** 
Precipitation  0.89490  0.44627 0.2088  0.001 ***
---
Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Permutation: free
Number of permutations: 999

Componentes principales

Enfermedades respiratorias_Year [biplot - Mes]

Unconstrained_Ambiente

Unconstrained_Contaminantes

Unconstrained_Contaminantes

Este es el método más recomendado según Blanchet, Legendre & Borcard. 2008. Ecology 89, 2623-2623 para seleccionar las variables del modelo.

upr <- vegan::cca(M1[,4:8] ~ . , data=M2[,c(-1)]) # Full Model
lwr <- vegan::cca(M1[,4:8] ~ 1 , data=M2[,c(-1)]) # Null model
## Stepwise Selection via Adjusted R^2
### 1. Global test
### 2. Radjusted forward selection

finalModel <- ordiR2step(lwr, upr, trace = FALSE)

finalModel

# VIF 
 #finalModel <- mods2

vif.cca(finalModel) # estamos "OK"!

De todas las variables ambientales y las de contaminantes, solo el SO2 y el PM2.5 mostraron un asociación significativa con la lista completa de enfermedades respiratorias (i.e, J:00, J:02, J:03, J:06 y J:20). Esto no descarta que algún contaminante o variable ambiental (o cualquiera de sus combinaciones) se asocien específicatmente con alguna enfermedad en particular.Habrá que ver esto con un análisis univariado (i.e.,incluso regresiones múltiples)

Permutation test for cca under reduced model
Terms added sequentially (first to last)
Permutation: free
Number of permutations: 999

Model: cca(formula = M1[, 4:8] ~ SO2 + PM2.5, data = M2[, -1])
         Df ChiSquare       F Pr(>F)    
SO2       1 0.0031738 10.0912  0.001 ***
PM2.5     1 0.0009931  3.1577  0.038 *  
Residual 62 0.0194997                   
---
Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

RDA Triplot

Enfermedades respiratorias_Year [biplot - Mes]

Unconstrained_Ambiente

Este es el método más recomendado según Blanchet, Legendre & Borcard. 2008. Ecology 89, 2623-2623 para seleccionar las variables del modelo.

upr <- vegan::cca(M1[,4:7] ~ . , data=M2[,-1]) # Full Model
lwr <- vegan::cca(M1[,4:7] ~ 1 , data=M2[,-1]) # Null model
## Stepwise Selection via Adjusted R^2
### 1. Global test
### 2. Radjusted forward selection

finalModel <- ordiR2step(lwr, upr, trace = FALSE)

finalModel

# VIF 
 #finalModel <- mods2

vif.cca(finalModel) # estamos "OK"!

** De todas las variables ambientales y las de contaminantes, solo las variables que se muestran abajo mostraron un asociación significativa con la lista completa de enfermedades AGUDAS (i.e, J:01, J:04, J:21 y J:22). Esto no descarta que algún contaminante o variable ambiental (o cualquiera de sus combinaciones) se asocien específicatmente con alguna enfermedad en particular.Habrá que ver esto con un análisis univariado (i.e.,incluso regresiones múltiples)**

Permutation test for cca under reduced model
Terms added sequentially (first to last)
Permutation: free
Number of permutations: 999

Model: cca(formula = M1[, 4:7] ~ Temp.min, data = M2[, -1])
         Df ChiSquare      F Pr(>F)  
Temp.min  1  0.015518 3.7399  0.044 *
Residual 63  0.261416                
---
Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

RDA Triplot

Enfermedades respiratorias_Year [biplot - Mes]

Unconstrained_Ambiente

Enfermedades respiratorias_Year [biplot - Mes]

Unconstrained_Ambiente

Este es el método más recomendado según Blanchet, Legendre & Borcard. 2008. Ecology 89, 2623-2623 para seleccionar las variables del modelo.

M1 <- decostand(M1[,4:12], method = "hellinger")
#M2 <- as_tibble(M2)
upr <- cca(M1 ~ . , data= M2[,c(-1)]) # Full Model
lwr <- cca(M1 ~ 1, data=M2[,c(-1)]) # Null model
## Stepwise Selection via Adjusted R^2
### 1. Global test
### 2. Radjusted forward selection

finalModel <- ordiR2step(lwr, upr, trace = TRUE,
                         permutations = 100)

finalModel

# VIF 
 #finalModel <- mods2

vif.cca(finalModel) # estamos "OK"!

** De todas las variables ambientales y las de contaminantes, solo las variables que se muestran abajo mostraron un asociación significativa con la lista completa de enfermedades AGUDAS (i.e, J:01, J:04, J:21 y J:22). Esto no descarta que algún contaminante o variable ambiental (o cualquiera de sus combinaciones) se asocien específicatmente con alguna enfermedad en particular.Habrá que ver esto con un análisis univariado (i.e.,incluso regresiones múltiples)**

Call: cca(formula = M1. ~ SO2 + Temp.min + Precipitation, data = M2[, -1])

-- Model Summary --

              Inertia Proportion Rank
Total         0.02152    1.00000     
Constrained   0.00347    0.16124    3
Unconstrained 0.01805    0.83876    8

Inertia is scaled Chi-square

-- Eigenvalues --

Eigenvalues for constrained axes:
     CCA1      CCA2      CCA3 
0.0018414 0.0015061 0.0001222 

Eigenvalues for unconstrained axes:
     CA1      CA2      CA3      CA4      CA5      CA6      CA7      CA8 
0.007702 0.004589 0.002365 0.001178 0.000900 0.000624 0.000431 0.000260 

Permutation test for cca under reduced model
Terms added sequentially (first to last)
Permutation: free
Number of permutations: 999

Model: cca(formula = M1. ~ SO2 + Temp.min + Precipitation, data = M2[, -1])
              Df ChiSquare      F Pr(>F)    
SO2            1 0.0017743 5.9964  0.001 ***
Temp.min       1 0.0008432 2.8495  0.038 *  
Precipitation  1 0.0008523 2.8803  0.034 *  
Residual      61 0.0180497                  
---
Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

RDA Triplot

Enfermedades respiratorias_Year [biplot - Year]

Unconstrained_Ambiente

Enfermedades respiratorias_Year [biplot - Year]

Unconstrained_Ambiente

Este es el método más recomendado según Blanchet, Legendre & Borcard. 2008. Ecology 89, 2623-2623 para seleccionar las variables del modelo.

upr <- vegan::cca(M1[,4:7] ~ . , data=M2[,c(-1)]) # Full Model
lwr <- vegan::cca(M1[,4:7] ~ 1 , data=M2[,c(-1)]) # Null model
## Stepwise Selection via Adjusted R^2
### 1. Global test
### 2. Radjusted forward selection

finalModel <- ordiR2step(lwr, upr, trace = FALSE)

finalModel

# VIF 
 #finalModel <- mods2

vif.cca(finalModel) # estamos "OK"!

** De todas las variables ambientales y las de contaminantes, solo las variables que se muestran abajo mostraron un asociación significativa con la lista completa de enfermedades AGUDAS (i.e, J:01, J:04, J:21 y J:22). Esto no descarta que algún contaminante o variable ambiental (o cualquiera de sus combinaciones) se asocien específicatmente con alguna enfermedad en particular.Habrá que ver esto con un análisis univariado (i.e.,incluso regresiones múltiples)**

Permutation test for cca under reduced model
Terms added sequentially (first to last)
Permutation: free
Number of permutations: 999

Model: cca(formula = M1[, 4:7] ~ SO2, data = M2[, -1])
         Df ChiSquare      F Pr(>F)  
SO2       1 0.0014445 4.1296  0.032 *
Residual 63 0.0220369                
---
Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

RDA Triplot

Componente.Princ 1 [J:00, J:02, J:03, J:20]

Call:
glm(formula = enf.PC1 ~ PM2.5 + I(PM2.5^2), data = m1)

Coefficients:
              Estimate Std. Error t value Pr(>|t|)  
(Intercept) -3.2781656  1.3924177  -2.354   0.0207 *
PM2.5        0.1643916  0.0707806   2.323   0.0224 *
I(PM2.5^2)  -0.0017726  0.0008344  -2.124   0.0363 *
---
Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

(Dispersion parameter for gaussian family taken to be 4.413727)

    Null deviance: 435.56  on 95  degrees of freedom
Residual deviance: 410.48  on 93  degrees of freedom
AIC: 419.92

Number of Fisher Scoring iterations: 2

Modelo Unimodal: Componente 1 [J:00, J:02, J:03, J:20] ~ PM2.5

Componente.Princ 1 [J:01, J:04, J:06, J:21, J:22]

Call:
glm(formula = enf.PC2 ~ SO2 + I(SO2^2), data = m1)

Coefficients:
             Estimate Std. Error t value Pr(>|t|)    
(Intercept) -0.897037   0.256142  -3.502 0.000854 ***
SO2          0.371863   0.094728   3.926 0.000217 ***
I(SO2^2)    -0.017976   0.006127  -2.934 0.004668 ** 
---
Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

(Dispersion parameter for gaussian family taken to be 0.9755507)

    Null deviance: 83.373  on 65  degrees of freedom
Residual deviance: 61.460  on 63  degrees of freedom
  (30 observations deleted due to missingness)
AIC: 190.6

Number of Fisher Scoring iterations: 2

Modelo Unimodal: Componente 1 [J:01, J:04, J:06, J:21, J:22] ~ SO2

J:01

Call:
glm(formula = J01 ~ SO2 + NO2 + I(NO2^2), family = poisson, data = m1)

Coefficients:
              Estimate Std. Error z value Pr(>|z|)    
(Intercept)  2.6566437  0.2508429  10.591  < 2e-16 ***
SO2          0.0132218  0.0051738   2.556  0.01060 *  
NO2          0.0427312  0.0133555   3.200  0.00138 ** 
I(NO2^2)    -0.0005617  0.0001726  -3.255  0.00113 ** 
---
Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

(Dispersion parameter for poisson family taken to be 1)

    Null deviance: 458.28  on 64  degrees of freedom
Residual deviance: 437.09  on 61  degrees of freedom
  (31 observations deleted due to missingness)
AIC: 784.3

Number of Fisher Scoring iterations: 4

Modelo Lineal: J:01 ~ SO2

Modelo unimodal: J:01 ~ NO2

J:04

Call:
glm(formula = J04 ~ SO2 + NO2 + I(NO2^2), family = poisson, data = m1)

Coefficients:
              Estimate Std. Error z value Pr(>|z|)    
(Intercept)  2.8267550  0.2350162  12.028  < 2e-16 ***
SO2          0.0155360  0.0048493   3.204  0.00136 ** 
NO2          0.0392781  0.0125245   3.136  0.00171 ** 
I(NO2^2)    -0.0005171  0.0001619  -3.194  0.00140 ** 
---
Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

(Dispersion parameter for poisson family taken to be 1)

    Null deviance: 319.02  on 64  degrees of freedom
Residual deviance: 293.62  on 61  degrees of freedom
  (31 observations deleted due to missingness)
AIC: 650.45

Number of Fisher Scoring iterations: 4

Modelo Lineal: J:04 ~ SO2

Modelo unimodal: J:04 ~ NO2

J:02

Call:
glm(formula = J02 ~ PM2.5 + I(PM2.5^2) + SO2 + NO2 + I(NO2^2), 
    family = poisson, data = m1)

Coefficients:
              Estimate Std. Error z value Pr(>|z|)    
(Intercept)  5.032e+00  5.508e-02   91.36   <2e-16 ***
PM2.5        3.800e-02  1.893e-03   20.07   <2e-16 ***
I(PM2.5^2)  -5.531e-04  2.456e-05  -22.52   <2e-16 ***
SO2          1.196e-02  1.017e-03   11.76   <2e-16 ***
NO2          6.181e-02  2.806e-03   22.03   <2e-16 ***
I(NO2^2)    -8.154e-04  3.606e-05  -22.61   <2e-16 ***
---
Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

(Dispersion parameter for poisson family taken to be 1)

    Null deviance: 11995  on 64  degrees of freedom
Residual deviance: 10490  on 59  degrees of freedom
  (31 observations deleted due to missingness)
AIC: 11054

Number of Fisher Scoring iterations: 4

Modelo Lineal: J:02 ~ SO2

Modelo unimodal: J:02 ~ NO2

Modelo unimodal: J:02 ~ PM2.5

Componente.Princ 1 [J:00, J:02, J:03, J:20]

Call:
glm(formula = enf.PC1 ~ NO2 + I(NO2^2), data = m1)

Coefficients:
             Estimate Std. Error t value Pr(>|t|)  
(Intercept) -5.168226   1.979610  -2.611   0.0106 *
NO2          0.277979   0.109377   2.541   0.0127 *
I(NO2^2)    -0.003436   0.001448  -2.372   0.0198 *
---
Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

(Dispersion parameter for gaussian family taken to be 4.399807)

    Null deviance: 430.78  on 93  degrees of freedom
Residual deviance: 400.38  on 91  degrees of freedom
  (2 observations deleted due to missingness)
AIC: 410.98

Number of Fisher Scoring iterations: 2

Modelo Unimodal: Componente 1 [J:00, J:02, J:03, J:20] ~ NO2

Componente.Princ 1 [J:00, J:02, J:03, J:20]

Call:
glm(formula = enf.PC1 ~ SO2, data = m1)

Coefficients:
            Estimate Std. Error t value Pr(>|t|)  
(Intercept)  0.67259    0.39458   1.705   0.0931 .
SO2         -0.03458    0.06230  -0.555   0.5808  
---
Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

(Dispersion parameter for gaussian family taken to be 4.573962)

    Null deviance: 294.14  on 65  degrees of freedom
Residual deviance: 292.73  on 64  degrees of freedom
  (30 observations deleted due to missingness)
AIC: 291.61

Number of Fisher Scoring iterations: 2

Call:
glm(formula = SO2 ~ NO2, data = m1)

Coefficients:
            Estimate Std. Error t value Pr(>|t|)    
(Intercept)  9.13790    1.92131   4.756 1.19e-05 ***
NO2         -0.11972    0.04944  -2.421   0.0184 *  
---
Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

(Dispersion parameter for gaussian family taken to be 16.8442)

    Null deviance: 1159.9  on 64  degrees of freedom
Residual deviance: 1061.2  on 63  degrees of freedom
  (31 observations deleted due to missingness)
AIC: 371.99

Number of Fisher Scoring iterations: 2

Modelo Unimodal: Componente 1 [J:00, J:02, J:03, J:20] ~ SO2

Enf. Respiratoria J:21

Call:
glm(formula = J21 ~ SO2 + I(SO2^2) + NO2 + I(NO2^2), family = poisson, 
    data = m1)

Coefficients:
              Estimate Std. Error z value Pr(>|z|)    
(Intercept)  1.2168146  0.3071446   3.962 7.44e-05 ***
SO2          0.2313072  0.0221371  10.449  < 2e-16 ***
I(SO2^2)    -0.0141242  0.0014384  -9.820  < 2e-16 ***
NO2          0.0671286  0.0155881   4.306 1.66e-05 ***
I(NO2^2)    -0.0007123  0.0002003  -3.556 0.000377 ***
---
Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

(Dispersion parameter for poisson family taken to be 1)

    Null deviance: 920.86  on 64  degrees of freedom
Residual deviance: 795.95  on 60  degrees of freedom
  (31 observations deleted due to missingness)
AIC: 1112.3

Number of Fisher Scoring iterations: 5

Modelo Unimodal: J:21 ~ SO2

Modelo Unimodal: J:21 ~ NO2

[1] "Peak of cases NO2 = 43.8 μg/m3"

Enf. Respiratoria J:22

Call:
glm(formula = J22 ~ PM2.5 + I(PM2.5^2) + SO2 + I(SO2^2), family = poisson, 
    data = m1)

Coefficients:
              Estimate Std. Error z value Pr(>|z|)    
(Intercept)  0.6957071  0.2486214   2.798  0.00514 ** 
PM2.5        0.0863751  0.0115707   7.465 8.33e-14 ***
I(PM2.5^2)  -0.0008477  0.0001381  -6.137 8.39e-10 ***
SO2          0.1669453  0.0223394   7.473 7.83e-14 ***
I(SO2^2)    -0.0093895  0.0014382  -6.529 6.63e-11 ***
---
Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

(Dispersion parameter for poisson family taken to be 1)

    Null deviance: 1115.78  on 65  degrees of freedom
Residual deviance:  978.62  on 61  degrees of freedom
  (30 observations deleted due to missingness)
AIC: 1276.8

Number of Fisher Scoring iterations: 5

Modelo Unimodal: J:22 ~ PM2.5

Modelo Unimodal: J:22 ~ SO2

[1] "Peak of cases SO2 = 8.98 μg/m3"

Enf. Respiratoria [J:00]

    Shapiro-Wilk normality test

data:  m1$J0
W = 0.98371, p-value = 0.2807

Call:
glm(formula = J0 ~ PM2.5 + I(PM2.5^2), data = m1)

Coefficients:
             Estimate Std. Error t value Pr(>|t|)    
(Intercept) 2069.6765   606.8088   3.411  0.00096 ***
PM2.5         78.3261    30.8458   2.539  0.01277 *  
I(PM2.5^2)    -0.8535     0.3636  -2.347  0.02104 *  
---
Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

(Dispersion parameter for gaussian family taken to be 838243.5)

    Null deviance: 83558974  on 95  degrees of freedom
Residual deviance: 77956648  on 93  degrees of freedom
AIC: 1586.7

Number of Fisher Scoring iterations: 2

Modelo Unimodal: J:00 ~ PM2.5

Enf. Respiratoria J:02

    Shapiro-Wilk normality test

data:  m1$J03
W = 0.94716, p-value = 0.000723

Call:
glm(formula = J03 ~ PM2.5 + I(PM2.5^2), family = poisson, data = m1)

Coefficients:
              Estimate Std. Error z value Pr(>|z|)    
(Intercept)  6.818e+00  1.704e-02  400.00   <2e-16 ***
PM2.5        3.198e-02  8.580e-04   37.28   <2e-16 ***
I(PM2.5^2)  -3.370e-04  1.013e-05  -33.27   <2e-16 ***
---
Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

(Dispersion parameter for poisson family taken to be 1)

    Null deviance: 24790  on 95  degrees of freedom
Residual deviance: 23179  on 93  degrees of freedom
AIC: 24072

Number of Fisher Scoring iterations: 4

Modelo Unimodal: J:02 ~ PM2.5

Enf. Respiratoria [J:00]

    Shapiro-Wilk normality test

data:  m1$J0
W = 0.98371, p-value = 0.2807

Call:
glm(formula = J0 ~ NO2 + I(NO2^2), data = m1)

Coefficients:
             Estimate Std. Error t value Pr(>|t|)   
(Intercept) 1192.6046   865.9138   1.377  0.17180   
NO2          133.1846    47.8431   2.784  0.00653 **
I(NO2^2)      -1.6757     0.6335  -2.645  0.00962 **
---
Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

(Dispersion parameter for gaussian family taken to be 841828.5)

    Null deviance: 83328334  on 93  degrees of freedom
Residual deviance: 76606390  on 91  degrees of freedom
  (2 observations deleted due to missingness)
AIC: 1554.2

Number of Fisher Scoring iterations: 2

Modelo Unimodal: J:00 ~ PM2.5

Enf. Respiratoria J:02

    Shapiro-Wilk normality test

data:  m1$J02
W = 0.93817, p-value = 0.0002055

Call:
glm(formula = J02 ~ NO2 + I(NO2^2), family = poisson, data = m1)

Coefficients:
              Estimate Std. Error z value Pr(>|z|)    
(Intercept)  5.792e+00  3.727e-02  155.41   <2e-16 ***
NO2          4.854e-02  2.046e-03   23.72   <2e-16 ***
I(NO2^2)    -6.252e-04  2.708e-05  -23.08   <2e-16 ***
---
Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

(Dispersion parameter for poisson family taken to be 1)

    Null deviance: 16714  on 93  degrees of freedom
Residual deviance: 16113  on 91  degrees of freedom
  (2 observations deleted due to missingness)
AIC: 16907

Number of Fisher Scoring iterations: 4

Modelo Unimodal: J:02 ~ PM2.5

Enf. Respiratoria J:03

    Shapiro-Wilk normality test

data:  m1$J03
W = 0.94716, p-value = 0.000723

Call:
glm(formula = J03 ~ NO2 + I(NO2^2), family = poisson, data = m1)

Coefficients:
              Estimate Std. Error z value Pr(>|z|)    
(Intercept)  6.701e+00  2.461e-02  272.26   <2e-16 ***
NO2          4.393e-02  1.355e-03   32.41   <2e-16 ***
I(NO2^2)    -5.795e-04  1.799e-05  -32.21   <2e-16 ***
---
Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

(Dispersion parameter for poisson family taken to be 1)

    Null deviance: 24321  on 93  degrees of freedom
Residual deviance: 23204  on 91  degrees of freedom
  (2 observations deleted due to missingness)
AIC: 24078

Number of Fisher Scoring iterations: 4

Modelo Unimodal: J:03 ~ PM2.5

Enf. Respiratoria J:06

Call:
glm(formula = J06 ~ SO2 + +NO2 + I(NO2^2), data = m1)

Coefficients:
             Estimate Std. Error t value Pr(>|t|)    
(Intercept) -308.2729   233.9980  -1.317 0.192626    
SO2          -25.2919     5.4470  -4.643 1.87e-05 ***
NO2           48.8569    12.3498   3.956 0.000201 ***
I(NO2^2)      -0.6019     0.1580  -3.811 0.000325 ***
---
Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

(Dispersion parameter for gaussian family taken to be 30944.81)

    Null deviance: 3068443  on 64  degrees of freedom
Residual deviance: 1887633  on 61  degrees of freedom
  (31 observations deleted due to missingness)
AIC: 862.43

Number of Fisher Scoring iterations: 2

Call:
glm(formula = J06 ~ NO2 + I(NO2^2), data = m1)

Coefficients:
            Estimate Std. Error t value Pr(>|t|)  
(Intercept)    2.699    203.611   0.013   0.9895  
NO2           23.441     11.250   2.084   0.0400 *
I(NO2^2)      -0.269      0.149  -1.806   0.0742 .
---
Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

(Dispersion parameter for gaussian family taken to be 46545.41)

    Null deviance: 4516085  on 93  degrees of freedom
Residual deviance: 4235632  on 91  degrees of freedom
  (2 observations deleted due to missingness)
AIC: 1282

Number of Fisher Scoring iterations: 2

NO2 - SO2

Modelo Unimodal: J:06 ~ NO2

Modelo Unimodal: J:06 ~ PM2.5

Enf. Respiratoria J:20

Call:
glm(formula = J20 ~ SO2 + NO2 + I(NO2^2), family = poisson, data = m1)

Coefficients:
              Estimate Std. Error z value Pr(>|z|)    
(Intercept)  4.830e+00  7.013e-02   68.86   <2e-16 ***
SO2         -2.427e-02  1.514e-03  -16.03   <2e-16 ***
NO2          7.291e-02  3.631e-03   20.08   <2e-16 ***
I(NO2^2)    -8.731e-04  4.589e-05  -19.03   <2e-16 ***
---
Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

(Dispersion parameter for poisson family taken to be 1)

    Null deviance: 3751.1  on 64  degrees of freedom
Residual deviance: 2984.2  on 61  degrees of freedom
  (31 observations deleted due to missingness)
AIC: 3507

Number of Fisher Scoring iterations: 4

Call:
glm(formula = J20 ~ NO2 + I(NO2^2), family = poisson, data = m1)

Coefficients:
              Estimate Std. Error z value Pr(>|z|)    
(Intercept)  5.209e+00  4.991e-02  104.37   <2e-16 ***
NO2          4.310e-02  2.709e-03   15.91   <2e-16 ***
I(NO2^2)    -4.900e-04  3.543e-05  -13.83   <2e-16 ***
---
Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

(Dispersion parameter for poisson family taken to be 1)

    Null deviance: 5425.2  on 93  degrees of freedom
Residual deviance: 5045.2  on 91  degrees of freedom
  (2 observations deleted due to missingness)
AIC: 5788.6

Number of Fisher Scoring iterations: 4

NO2 - SO2

Modelo Unimodal: J:20 ~ NO2

Enf. Respiratoria J:21

    Shapiro-Wilk normality test

data:  m1$J21
W = 0.87589, p-value = 1.935e-07

Call:
glm(formula = J21 ~ SO2 + I(SO2^2) + NO2 + I(NO2^2), family = poisson, 
    data = m1)

Coefficients:
              Estimate Std. Error z value Pr(>|z|)    
(Intercept)  1.2168146  0.3071446   3.962 7.44e-05 ***
SO2          0.2313072  0.0221371  10.449  < 2e-16 ***
I(SO2^2)    -0.0141242  0.0014384  -9.820  < 2e-16 ***
NO2          0.0671286  0.0155881   4.306 1.66e-05 ***
I(NO2^2)    -0.0007123  0.0002003  -3.556 0.000377 ***
---
Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

(Dispersion parameter for poisson family taken to be 1)

    Null deviance: 920.86  on 64  degrees of freedom
Residual deviance: 795.95  on 60  degrees of freedom
  (31 observations deleted due to missingness)
AIC: 1112.3

Number of Fisher Scoring iterations: 5

Call:
glm(formula = J21 ~ NO2 + I(NO2^2), family = poisson, data = m1)

Coefficients:
              Estimate Std. Error z value Pr(>|z|)    
(Intercept)  1.5930615  0.2331330   6.833 8.30e-12 ***
NO2          0.0769173  0.0125110   6.148 7.85e-10 ***
I(NO2^2)    -0.0008780  0.0001623  -5.411 6.28e-08 ***
---
Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

(Dispersion parameter for poisson family taken to be 1)

    Null deviance: 1228.3  on 93  degrees of freedom
Residual deviance: 1170.7  on 91  degrees of freedom
  (2 observations deleted due to missingness)
AIC: 1613.9

Number of Fisher Scoring iterations: 5

Modelo Unimodal: J:21 ~ NO2

Modelo Unimodal: J:21 ~ SO2

Modelo Unimodal: J:21 ~ SO2

Componente.Princ 1 [J0,J02,J03, J20]

Call:
glm(formula = enf.PC1 ~ amb.PCA.1, data = m1)

Coefficients:
              Estimate Std. Error t value Pr(>|t|)  
(Intercept) -7.564e-17  2.154e-01   0.000    1.000  
amb.PCA.1    2.574e-01  1.330e-01   1.935    0.056 .
---
Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

(Dispersion parameter for gaussian family taken to be 4.45605)

    Null deviance: 435.56  on 95  degrees of freedom
Residual deviance: 418.87  on 94  degrees of freedom
AIC: 419.86

Number of Fisher Scoring iterations: 2

Modelo Unimodal: Componente 1 [J01, J04,J06,J:21,J:22] ~ PM2.5

Componente.Princ.2[J01,J04,J06,J:21,J:22]

Call:
glm(formula = enf.PC2 ~ amb.PCA.1, data = m1)

Coefficients:
             Estimate Std. Error t value Pr(>|t|)
(Intercept) 4.884e-17  1.125e-01   0.000    1.000
amb.PCA.1   7.524e-02  6.942e-02   1.084    0.281

(Dispersion parameter for gaussian family taken to be 1.214156)

    Null deviance: 115.56  on 95  degrees of freedom
Residual deviance: 114.13  on 94  degrees of freedom
AIC: 295.04

Number of Fisher Scoring iterations: 2

Modelo Unimodal: Componente.Princ.2[J01,J04,J06,J:21,J:22] ~ amb.PCA.1

Predictor: AMB.CP.2[Hum., Precip.Pluvial]

Call:
glm(formula = enf.PC2 ~ amb.PCA.2 + SO2 + I(SO2^2), data = m1)

Coefficients:
             Estimate Std. Error t value Pr(>|t|)    
(Intercept) -0.841206   0.242933  -3.463 0.000975 ***
amb.PCA.2   -0.239040   0.082105  -2.911 0.004994 ** 
SO2          0.372587   0.089563   4.160 9.97e-05 ***
I(SO2^2)    -0.019330   0.005812  -3.326 0.001485 ** 
---
Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

(Dispersion parameter for gaussian family taken to be 0.8720619)

    Null deviance: 83.373  on 65  degrees of freedom
Residual deviance: 54.068  on 62  degrees of freedom
  (30 observations deleted due to missingness)
AIC: 184.14

Number of Fisher Scoring iterations: 2

Modelo: Componente 1 [J:00, J:01, J:02, J:03, J:20] ~ Amb.CP.2[Hum.Precip.Pluv]

Componente.Princ 1 [J:00,J:02,J:03,J:20]

Call:
glm(formula = enf.PC1 ~ cnrd.pca.1, data = m1)

Coefficients:
              Estimate Std. Error t value Pr(>|t|)
(Intercept) -1.782e-16  2.196e-01   0.000    1.000
cnrd.pca.1  -7.058e-02  2.004e-01  -0.352    0.726

(Dispersion parameter for gaussian family taken to be 4.627479)

    Null deviance: 435.56  on 95  degrees of freedom
Residual deviance: 434.98  on 94  degrees of freedom
AIC: 423.49

Number of Fisher Scoring iterations: 2

Enfermedades y PM2.5

Enfermedades y Ambiente

CONRED Componentes Principales

***VECTORS

               PC1      PC2     r2 Pr(>r)    
con.masas -0.80006  0.59992 0.6989  0.001 ***
incendios -0.28550 -0.95838 0.1422  0.005 ** 
lluvias    0.89490  0.44627 0.2088  0.001 ***
---
Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Permutation: free
Number of permutations: 999

Call:
glm(formula = enf.PC1 ~ cnrd.pca.1 + cnrd.pca.2, data = m1)

Coefficients:
              Estimate Std. Error t value Pr(>|t|)
(Intercept) -1.785e-16  2.207e-01   0.000    1.000
cnrd.pca.1  -7.058e-02  2.015e-01  -0.350    0.727
cnrd.pca.2   2.997e-02  2.083e-01   0.144    0.886

(Dispersion parameter for gaussian family taken to be 4.676196)

    Null deviance: 435.56  on 95  degrees of freedom
Residual deviance: 434.89  on 93  degrees of freedom
AIC: 425.47

Number of Fisher Scoring iterations: 2

CONRED (conc.masas y lluvias)

CONRED (incendios)

CONRED (conc.masas y lluvias)

CONRED (incendios)

CONRED (conc.masas y lluvias)

CONRED (incendios)

J06

Call:
glm(formula = J06 ~ amb.PCA.1 + cnrd.pca.1 + I(cnrd.pca.1^2) + 
    cnrd.pca.2 + I(cnrd.pca.2^2) + PM2.5 + I(PM2.5^2) + NO2 + 
    I(NO2^2), family = poisson, data = m1)

Coefficients:
                  Estimate Std. Error z value Pr(>|z|)    
(Intercept)      4.948e+00  5.764e-02  85.842  < 2e-16 ***
amb.PCA.1        9.091e-02  3.848e-03  23.622  < 2e-16 ***
cnrd.pca.1      -8.984e-02  6.678e-03 -13.454  < 2e-16 ***
I(cnrd.pca.1^2)  2.402e-01  1.354e-02  17.744  < 2e-16 ***
cnrd.pca.2      -4.777e-01  2.675e-02 -17.857  < 2e-16 ***
I(cnrd.pca.2^2) -2.697e-01  1.474e-02 -18.294  < 2e-16 ***
PM2.5            3.459e-02  1.738e-03  19.900  < 2e-16 ***
I(PM2.5^2)      -3.522e-04  2.019e-05 -17.443  < 2e-16 ***
NO2              1.706e-02  2.847e-03   5.990  2.1e-09 ***
I(NO2^2)        -1.028e-04  3.778e-05  -2.722   0.0065 ** 
---
Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

(Dispersion parameter for poisson family taken to be 1)

    Null deviance: 9999.7  on 93  degrees of freedom
Residual deviance: 7835.6  on 84  degrees of freedom
  (2 observations deleted due to missingness)
AIC: 8594.1

Number of Fisher Scoring iterations: 4

Ambiente [Temperatura]

PM2.5

CONRED (conc.masas y lluvias)

CONRED (incendios)

J20

Call:
glm(formula = J20 ~ amb.PCA.1 + amb.PCA.2 + +PM2.5 + I(PM2.5^2) + 
    NO2 + I(NO2^2), family = poisson, data = m1)

Coefficients:
              Estimate Std. Error z value Pr(>|z|)    
(Intercept)  4.793e+00  5.799e-02  82.645  < 2e-16 ***
amb.PCA.1    4.106e-02  3.403e-03  12.067  < 2e-16 ***
amb.PCA.2    2.597e-02  3.840e-03   6.762 1.36e-11 ***
PM2.5        2.911e-02  1.756e-03  16.578  < 2e-16 ***
I(PM2.5^2)  -2.733e-04  2.021e-05 -13.526  < 2e-16 ***
NO2          2.840e-02  2.845e-03   9.979  < 2e-16 ***
I(NO2^2)    -2.747e-04  3.787e-05  -7.254 4.05e-13 ***
---
Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

(Dispersion parameter for poisson family taken to be 1)

    Null deviance: 5425.2  on 93  degrees of freedom
Residual deviance: 4476.4  on 87  degrees of freedom
  (2 observations deleted due to missingness)
AIC: 5227.9

Number of Fisher Scoring iterations: 4

AMB.PC.1 [Temperatura(-)]

AMB.CP.2[Precip.Pluvial-Hum.Relativa]

PM2.5

NO2

enf.PCA1 [J0,J02,J03,J20]

Start:  AIC=284.3
enf.PC1 ~ PM2.5 + I(PM2.5^2) + NO2 + I(NO2^2) + Temp.min + Precipitation + 
    Rel.Humidity

                Df Deviance    AIC
- Rel.Humidity   1   228.97 282.31
- PM2.5          1   232.50 283.30
- I(PM2.5^2)     1   233.31 283.53
- Precipitation  1   235.69 284.19
<none>               228.93 284.30
- Temp.min       1   248.83 287.72
- NO2            1   262.58 291.21
- I(NO2^2)       1   263.59 291.46

Step:  AIC=282.31
enf.PC1 ~ PM2.5 + I(PM2.5^2) + NO2 + I(NO2^2) + Temp.min + Precipitation

                Df Deviance    AIC
- PM2.5          1   232.60 281.33
- I(PM2.5^2)     1   233.35 281.54
<none>               228.97 282.31
- Precipitation  1   240.83 283.59
+ Rel.Humidity   1   228.93 284.30
- Temp.min       1   248.97 285.75
- NO2            1   262.62 289.22
- I(NO2^2)       1   263.65 289.48

Step:  AIC=281.33
enf.PC1 ~ I(PM2.5^2) + NO2 + I(NO2^2) + Temp.min + Precipitation

                Df Deviance    AIC
- I(PM2.5^2)     1   233.65 279.62
<none>               232.60 281.33
+ PM2.5          1   228.97 282.31
- Precipitation  1   243.66 282.35
+ Rel.Humidity   1   232.50 283.30
- Temp.min       1   255.70 285.49
- NO2            1   276.04 290.46
- I(NO2^2)       1   276.47 290.56

Step:  AIC=279.62
enf.PC1 ~ NO2 + I(NO2^2) + Temp.min + Precipitation

                Df Deviance    AIC
<none>               233.65 279.62
- Precipitation  1   246.23 281.04
+ I(PM2.5^2)     1   232.60 281.33
+ PM2.5          1   233.35 281.54
+ Rel.Humidity   1   233.65 281.62
- Temp.min       1   257.00 283.82
- NO2            1   276.05 288.46
- I(NO2^2)       1   276.50 288.57

Call:
glm(formula = enf.PC1 ~ NO2 + I(NO2^2) + Temp.min + Precipitation, 
    data = m1.vars)

Coefficients:
               Estimate Std. Error t value Pr(>|t|)   
(Intercept)    0.004005   4.027751   0.001  0.99921   
NO2            0.461819   0.139961   3.300  0.00163 **
I(NO2^2)      -0.005979   0.001802  -3.317  0.00155 **
Temp.min      -0.520074   0.212398  -2.449  0.01728 * 
Precipitation  0.004596   0.002556   1.798  0.07726 . 
---
Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

(Dispersion parameter for gaussian family taken to be 3.894135)

    Null deviance: 292.37  on 64  degrees of freedom
Residual deviance: 233.65  on 60  degrees of freedom
AIC: 279.62

Number of Fisher Scoring iterations: 2

enf.PC1 ~ NO2 + I(NO2^2) + Temp.min + Precipitation

Recomiendo sacar Precipitation del modelo, debido a que contribuye solo un 9 % a explicar el número de casos.

Call:
glm(formula = enf.PC1 ~ NO2 + I(NO2^2) + Temp.min, data = m1.vars)

Coefficients:
             Estimate Std. Error t value Pr(>|t|)   
(Intercept) -2.006101   3.939567  -0.509  0.61244   
NO2          0.433898   0.141618   3.064  0.00325 **
I(NO2^2)    -0.005626   0.001824  -3.085  0.00306 **
Temp.min    -0.330934   0.187851  -1.762  0.08313 . 
---
Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

(Dispersion parameter for gaussian family taken to be 4.036595)

    Null deviance: 292.37  on 64  degrees of freedom
Residual deviance: 246.23  on 61  degrees of freedom
AIC: 281.03

Number of Fisher Scoring iterations: 2

J21

Start:  AIC=1201.46
J21 ~ NO2 + I(NO2^2) + PM2.5 + Precipitation + Rel.Humidity

                Df Deviance    AIC
- PM2.5          1   883.62 1200.0
- Precipitation  1   883.96 1200.3
<none>               883.12 1201.5
- NO2            1   892.10 1208.5
- I(NO2^2)       1   892.26 1208.6
- Rel.Humidity   1   896.29 1212.6

Step:  AIC=1199.96
J21 ~ NO2 + I(NO2^2) + Precipitation + Rel.Humidity

                Df Deviance    AIC
- Precipitation  1   884.52 1198.9
<none>               883.62 1200.0
- NO2            1   893.90 1208.2
- I(NO2^2)       1   894.16 1208.5
- Rel.Humidity   1   899.18 1213.5

Step:  AIC=1198.86
J21 ~ NO2 + I(NO2^2) + Rel.Humidity

               Df Deviance    AIC
<none>              884.52 1198.9
- NO2           1   894.93 1207.3
- I(NO2^2)      1   895.54 1207.9
- Rel.Humidity  1   907.42 1219.8

Call:
glm(formula = J21 ~ NO2 + I(NO2^2) + Rel.Humidity, family = "poisson", 
    data = m1.vars)

Coefficients:
               Estimate Std. Error z value Pr(>|z|)    
(Intercept)   4.2996066  0.5199626   8.269  < 2e-16 ***
NO2           0.0485994  0.0154659   3.142  0.00168 ** 
I(NO2^2)     -0.0006381  0.0001979  -3.224  0.00126 ** 
Rel.Humidity -0.0260668  0.0054639  -4.771 1.84e-06 ***
---
Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

(Dispersion parameter for poisson family taken to be 1)

    Null deviance: 920.86  on 64  degrees of freedom
Residual deviance: 884.52  on 61  degrees of freedom
AIC: 1198.9

Number of Fisher Scoring iterations: 5

###

J22

Call:
glm(formula = J22 ~ PM2.5 + I(PM2.5^2) + Temp.max + I(Temp.max^2) + 
    Precipitation + Rel.Humidity, family = "poisson", data = m1.vars)

Coefficients:
                Estimate Std. Error z value Pr(>|z|)    
(Intercept)   -1.402e+01  1.055e+01  -1.329  0.18380    
PM2.5          6.583e-02  1.131e-02   5.823 5.79e-09 ***
I(PM2.5^2)    -6.183e-04  1.330e-04  -4.647 3.37e-06 ***
Temp.max       1.342e+00  8.212e-01   1.634  0.10228    
I(Temp.max^2) -2.555e-02  1.590e-02  -1.606  0.10818    
Precipitation  1.111e-03  4.006e-04   2.774  0.00554 ** 
Rel.Humidity  -2.843e-02  1.158e-02  -2.455  0.01410 *  
---
Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

(Dispersion parameter for poisson family taken to be 1)

    Null deviance: 1092.9  on 64  degrees of freedom
Residual deviance: 1003.4  on 58  degrees of freedom
AIC: 1302.5

Number of Fisher Scoring iterations: 5

J06

Call:
glm(formula = J06 ~ NO2 + I(NO2^2) + PM2.5 + I(PM2.5^2) + Temp.min + 
    Precipitation + Rel.Humidity, family = "poisson", data = m1.vars)

Coefficients:
                Estimate Std. Error z value Pr(>|z|)    
(Intercept)    4.108e+00  1.822e-01  22.547  < 2e-16 ***
NO2            9.846e-02  3.981e-03  24.731  < 2e-16 ***
I(NO2^2)      -1.114e-03  5.009e-05 -22.248  < 2e-16 ***
PM2.5          3.864e-02  2.474e-03  15.621  < 2e-16 ***
I(PM2.5^2)    -4.460e-04  3.048e-05 -14.636  < 2e-16 ***
Temp.min      -2.960e-02  4.963e-03  -5.964 2.46e-09 ***
Precipitation  1.144e-03  8.321e-05  13.753  < 2e-16 ***
Rel.Humidity  -4.348e-03  1.920e-03  -2.264   0.0236 *  
---
Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

(Dispersion parameter for poisson family taken to be 1)

    Null deviance: 6338.2  on 64  degrees of freedom
Residual deviance: 4645.9  on 57  degrees of freedom
AIC: 5177.7

Number of Fisher Scoring iterations: 4

J01

Call:
glm(formula = J01 ~ NO2 + I(NO2^2) + Temp.min + Precipitation, 
    family = "poisson", data = m1.vars)

Coefficients:
                Estimate Std. Error z value Pr(>|z|)    
(Intercept)    3.7330424  0.3619455  10.314  < 2e-16 ***
NO2            0.0501492  0.0134074   3.740 0.000184 ***
I(NO2^2)      -0.0007434  0.0001746  -4.258 2.06e-05 ***
Temp.min      -0.0609682  0.0184989  -3.296 0.000981 ***
Precipitation -0.0005102  0.0002361  -2.161 0.030691 *  
---
Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

(Dispersion parameter for poisson family taken to be 1)

    Null deviance: 458.28  on 64  degrees of freedom
Residual deviance: 413.60  on 60  degrees of freedom
AIC: 762.82

Number of Fisher Scoring iterations: 5

J04

Start:  AIC=658.44
J04 ~ NO2 + I(NO2^2)

           Df Deviance    AIC
<none>          303.62 658.44
- NO2       1   315.88 668.70
- I(NO2^2)  1   317.68 670.50

Call:
glm(formula = J04 ~ NO2 + I(NO2^2), family = "poisson", data = m1.vars)

Coefficients:
              Estimate Std. Error z value Pr(>|z|)    
(Intercept)  2.8760684  0.2351118  12.233  < 2e-16 ***
NO2          0.0427341  0.0125017   3.418 0.000630 ***
I(NO2^2)    -0.0005863  0.0001608  -3.647 0.000265 ***
---
Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

(Dispersion parameter for poisson family taken to be 1)

    Null deviance: 319.02  on 64  degrees of freedom
Residual deviance: 303.62  on 62  degrees of freedom
AIC: 658.44

Number of Fisher Scoring iterations: 4

J0

Start:  AIC=11244.8
J0 ~ NO2 + I(NO2^2) + PM2.5 + I(PM2.5^2) + Temp.min + Precipitation + 
    Rel.Humidity

                Df Deviance   AIC
<none>                10575 11245
- Rel.Humidity   1    10662 11330
- PM2.5          1    10714 11382
- I(PM2.5^2)     1    10748 11417
- Precipitation  1    11157 11826
- Temp.min       1    11555 12224
- NO2            1    11866 12535
- I(NO2^2)       1    11984 12652

Call:
glm(formula = J0 ~ NO2 + I(NO2^2) + PM2.5 + I(PM2.5^2) + Temp.min + 
    Precipitation + Rel.Humidity, family = "poisson", data = m1.vars)

Coefficients:
                Estimate Std. Error z value Pr(>|z|)    
(Intercept)    8.586e+00  6.406e-02 134.033   <2e-16 ***
NO2            4.531e-02  1.286e-03  35.221   <2e-16 ***
I(NO2^2)      -6.080e-04  1.654e-05 -36.767   <2e-16 ***
PM2.5          9.685e-03  8.264e-04  11.720   <2e-16 ***
I(PM2.5^2)    -1.323e-04  1.016e-05 -13.023   <2e-16 ***
Temp.min      -5.513e-02  1.758e-03 -31.355   <2e-16 ***
Precipitation  7.198e-04  2.972e-05  24.216   <2e-16 ***
Rel.Humidity  -6.342e-03  6.792e-04  -9.338   <2e-16 ***
---
Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

(Dispersion parameter for poisson family taken to be 1)

    Null deviance: 13299  on 64  degrees of freedom
Residual deviance: 10575  on 57  degrees of freedom
AIC: 11245

Number of Fisher Scoring iterations: 4

NO2 ~ Temp.min

Start:  AIC=486.83
NO2 ~ Temp.min + Temp.max + Precipitation + Rel.Humidity

                Df Deviance    AIC
- Rel.Humidity   1   5663.3 484.84
- Precipitation  1   5681.6 485.05
- Temp.max       1   5702.4 485.29
<none>               5662.0 486.83
- Temp.min       1   5991.7 488.51

Step:  AIC=484.84
NO2 ~ Temp.min + Temp.max + Precipitation

                Df Deviance    AIC
- Precipitation  1   5687.8 483.12
- Temp.max       1   5744.6 483.77
<none>               5663.3 484.84
- Temp.min       1   6001.5 486.61
+ Rel.Humidity   1   5662.0 486.83

Step:  AIC=483.12
NO2 ~ Temp.min + Temp.max

                Df Deviance    AIC
- Temp.max       1   5747.6 481.80
<none>               5687.8 483.12
+ Precipitation  1   5663.3 484.84
+ Rel.Humidity   1   5681.6 485.05
- Temp.min       1   6422.6 489.02

Step:  AIC=481.8
NO2 ~ Temp.min

                Df Deviance    AIC
<none>               5747.6 481.80
+ Temp.max       1   5687.8 483.12
+ Rel.Humidity   1   5736.5 483.68
+ Precipitation  1   5744.6 483.77
- Temp.min       1   6889.9 491.58

Call:
glm(formula = NO2 ~ Temp.min, data = m1.vars)

Coefficients:
            Estimate Std. Error t value Pr(>|t|)    
(Intercept)  82.5151    12.7848   6.454 1.77e-08 ***
Temp.min     -2.8303     0.7998  -3.539 0.000762 ***
---
Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

(Dispersion parameter for gaussian family taken to be 91.23111)

    Null deviance: 6889.9  on 64  degrees of freedom
Residual deviance: 5747.6  on 63  degrees of freedom
AIC: 481.8

Number of Fisher Scoring iterations: 2

SO2 ~ Temp.max

Start:  AIC=373.23
SO2 ~ Temp.max + Temp.min + Precipitation + Rel.Humidity

                Df Deviance    AIC
- Temp.min       1   989.67 371.46
- Rel.Humidity   1   992.44 371.64
- Precipitation  1  1002.43 372.29
<none>               986.25 373.23
- Temp.max       1  1060.35 375.94

Step:  AIC=371.46
SO2 ~ Temp.max + Precipitation + Rel.Humidity

                Df Deviance    AIC
- Rel.Humidity   1   998.62 370.04
- Precipitation  1  1002.96 370.32
<none>               989.67 371.46
+ Temp.min       1   986.25 373.23
- Temp.max       1  1124.96 377.78

Step:  AIC=370.04
SO2 ~ Temp.max + Precipitation

                Df Deviance    AIC
- Precipitation  1  1002.99 368.33
<none>               998.62 370.04
+ Rel.Humidity   1   989.67 371.46
+ Temp.min       1   992.44 371.64
- Temp.max       1  1152.09 377.33

Step:  AIC=368.33
SO2 ~ Temp.max

                Df Deviance    AIC
<none>              1002.99 368.33
+ Precipitation  1   998.62 370.04
+ Temp.min       1  1002.52 370.29
+ Rel.Humidity   1  1002.96 370.32
- Temp.max       1  1159.93 375.77

Call:
glm(formula = SO2 ~ Temp.max, data = m1.vars)

Coefficients:
            Estimate Std. Error t value Pr(>|t|)   
(Intercept) -27.9182    10.3854  -2.688  0.00918 **
Temp.max      1.2548     0.3996   3.140  0.00257 **
---
Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

(Dispersion parameter for gaussian family taken to be 15.92049)

    Null deviance: 1159.9  on 64  degrees of freedom
Residual deviance: 1003.0  on 63  degrees of freedom
AIC: 368.33

Number of Fisher Scoring iterations: 2

SO2 ~ Temperature

NO2 ~ PM2.5

Start:  AIC=492.9
NO2 ~ PM2.5 + I(PM2.5^2)

             Df Deviance    AIC
<none>            6611.0 492.90
- PM2.5       1   6856.7 493.27
- I(PM2.5^2)  1   6886.6 493.55

Call:
glm(formula = NO2 ~ PM2.5 + I(PM2.5^2), data = m1.vars)

Coefficients:
             Estimate Std. Error t value Pr(>|t|)   
(Intercept) 25.052187   9.101634   2.752  0.00775 **
PM2.5        0.724517   0.477293   1.518  0.13410   
I(PM2.5^2)  -0.009510   0.005916  -1.608  0.11301   
---
Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

(Dispersion parameter for gaussian family taken to be 106.6296)

    Null deviance: 6889.9  on 64  degrees of freedom
Residual deviance: 6611.0  on 62  degrees of freedom
AIC: 492.9

Number of Fisher Scoring iterations: 2

SO2 ~ Temp.max

Start:  AIC=373.23
SO2 ~ Temp.max + Temp.min + Precipitation + Rel.Humidity

                Df Deviance    AIC
- Temp.min       1   989.67 371.46
- Rel.Humidity   1   992.44 371.64
- Precipitation  1  1002.43 372.29
<none>               986.25 373.23
- Temp.max       1  1060.35 375.94

Step:  AIC=371.46
SO2 ~ Temp.max + Precipitation + Rel.Humidity

                Df Deviance    AIC
- Rel.Humidity   1   998.62 370.04
- Precipitation  1  1002.96 370.32
<none>               989.67 371.46
+ Temp.min       1   986.25 373.23
- Temp.max       1  1124.96 377.78

Step:  AIC=370.04
SO2 ~ Temp.max + Precipitation

                Df Deviance    AIC
- Precipitation  1  1002.99 368.33
<none>               998.62 370.04
+ Rel.Humidity   1   989.67 371.46
+ Temp.min       1   992.44 371.64
- Temp.max       1  1152.09 377.33

Step:  AIC=368.33
SO2 ~ Temp.max

                Df Deviance    AIC
<none>              1002.99 368.33
+ Precipitation  1   998.62 370.04
+ Temp.min       1  1002.52 370.29
+ Rel.Humidity   1  1002.96 370.32
- Temp.max       1  1159.93 375.77

Call:
glm(formula = SO2 ~ Temp.max, data = m1.vars)

Coefficients:
            Estimate Std. Error t value Pr(>|t|)   
(Intercept) -27.9182    10.3854  -2.688  0.00918 **
Temp.max      1.2548     0.3996   3.140  0.00257 **
---
Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

(Dispersion parameter for gaussian family taken to be 15.92049)

    Null deviance: 1159.9  on 64  degrees of freedom
Residual deviance: 1003.0  on 63  degrees of freedom
AIC: 368.33

Number of Fisher Scoring iterations: 2

J30

Start:  AIC=1454.97
`J:30` ~ NO2 + I(NO2^2) + PM2.5 + I(PM2.5^2) + Temp.min + Precipitation + 
    Rel.Humidity

                Df Deviance    AIC
<none>               972.30 1455.0
- PM2.5          1   986.16 1466.8
- I(PM2.5^2)     1   986.54 1467.2
- Rel.Humidity   1   987.56 1468.2
- Precipitation  1  1074.13 1554.8
- I(NO2^2)       1  1108.53 1589.2
- Temp.min       1  1122.99 1603.7
- NO2            1  1129.09 1609.8

Call:
glm(formula = `J:30` ~ NO2 + I(NO2^2) + PM2.5 + I(PM2.5^2) + 
    Temp.min + Precipitation + Rel.Humidity, family = "poisson", 
    data = m1.vars)

Coefficients:
                Estimate Std. Error z value Pr(>|z|)    
(Intercept)    5.371e+00  2.930e-01  18.331  < 2e-16 ***
NO2            8.648e-02  7.150e-03  12.095  < 2e-16 ***
I(NO2^2)      -9.924e-04  8.787e-05 -11.294  < 2e-16 ***
PM2.5          1.956e-02  5.305e-03   3.688 0.000226 ***
I(PM2.5^2)    -2.341e-04  6.283e-05  -3.726 0.000195 ***
Temp.min      -8.906e-02  7.254e-03 -12.277  < 2e-16 ***
Precipitation  1.299e-03  1.280e-04  10.145  < 2e-16 ***
Rel.Humidity  -1.170e-02  2.992e-03  -3.912 9.15e-05 ***
---
Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

(Dispersion parameter for poisson family taken to be 1)

    Null deviance: 1316.3  on 64  degrees of freedom
Residual deviance:  972.3  on 57  degrees of freedom
AIC: 1455

Number of Fisher Scoring iterations: 4

J32

Start:  AIC=482.81
`J:32` ~ NO2 + I(NO2^2) + Temp.min

           Df Deviance    AIC
<none>          193.85 482.81
- NO2       1   203.51 490.47
- I(NO2^2)  1   206.06 493.02
- Temp.min  1   235.47 522.43

Call:
glm(formula = `J:32` ~ NO2 + I(NO2^2) + Temp.min, family = "poisson", 
    data = m1.vars)

Coefficients:
              Estimate Std. Error z value Pr(>|z|)    
(Intercept)  3.6693935  0.6226804   5.893 3.79e-09 ***
NO2          0.0830049  0.0277077   2.996 0.002738 ** 
I(NO2^2)    -0.0011797  0.0003528  -3.343 0.000827 ***
Temp.min    -0.1530024  0.0234398  -6.527 6.69e-11 ***
---
Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

(Dispersion parameter for poisson family taken to be 1)

    Null deviance: 241.40  on 64  degrees of freedom
Residual deviance: 193.85  on 61  degrees of freedom
AIC: 482.81

Number of Fisher Scoring iterations: 4

J44

Start:  AIC=389.38
`J:44` ~ NO2 + I(NO2^2) + PM2.5 + I(PM2.5^2)

             Df Deviance    AIC
<none>            126.28 389.38
- I(NO2^2)    1   129.27 390.38
- NO2         1   129.93 391.03
- I(PM2.5^2)  1   137.68 398.79
- PM2.5       1   138.10 399.21

Call:
glm(formula = `J:44` ~ NO2 + I(NO2^2) + PM2.5 + I(PM2.5^2), family = "poisson", 
    data = m1.vars)

Coefficients:
              Estimate Std. Error z value Pr(>|z|)   
(Intercept) -1.1252930  0.9128206  -1.233  0.21766   
NO2          0.0643187  0.0345842   1.860  0.06292 . 
I(NO2^2)    -0.0007178  0.0004256  -1.687  0.09170 . 
PM2.5        0.0970765  0.0299911   3.237  0.00121 **
I(PM2.5^2)  -0.0011389  0.0003625  -3.142  0.00168 **
---
Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

(Dispersion parameter for poisson family taken to be 1)

    Null deviance: 142.87  on 64  degrees of freedom
Residual deviance: 126.28  on 60  degrees of freedom
AIC: 389.38

Number of Fisher Scoring iterations: 5

Column ———— ###

J45

Start:  AIC=2705.25
`J:45` ~ NO2 + I(NO2^2) + PM2.5 + I(PM2.5^2) + Temp.min

             Df Deviance    AIC
<none>            2212.3 2705.2
- PM2.5       1   2215.8 2706.8
- I(PM2.5^2)  1   2221.6 2712.6
- Temp.min    1   2399.1 2890.1
- NO2         1   2431.7 2922.7
- I(NO2^2)    1   2442.6 2933.5

Call:
glm(formula = `J:45` ~ NO2 + I(NO2^2) + PM2.5 + I(PM2.5^2) + 
    Temp.min, family = "poisson", data = m1.vars)

Coefficients:
              Estimate Std. Error z value Pr(>|z|)    
(Intercept)  5.055e+00  1.800e-01  28.079  < 2e-16 ***
NO2          8.981e-02  6.296e-03  14.264  < 2e-16 ***
I(NO2^2)    -1.152e-03  7.910e-05 -14.561  < 2e-16 ***
PM2.5        9.003e-03  4.813e-03   1.871  0.06140 .  
I(PM2.5^2)  -1.755e-04  5.796e-05  -3.028  0.00246 ** 
Temp.min    -7.336e-02  5.337e-03 -13.744  < 2e-16 ***
---
Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

(Dispersion parameter for poisson family taken to be 1)

    Null deviance: 2617.3  on 64  degrees of freedom
Residual deviance: 2212.3  on 59  degrees of freedom
AIC: 2705.2

Number of Fisher Scoring iterations: 4

J18

Start:  AIC=3598.61
`J:18` ~ NO2 + I(NO2^2) + PM2.5 + I(PM2.5^2)

Call:
glm(formula = `J:18` ~ NO2 + I(NO2^2) + PM2.5 + I(PM2.5^2), family = "poisson", 
    data = m1.vars)

Coefficients:
              Estimate Std. Error z value Pr(>|z|)    
(Intercept)  3.288e+00  1.154e-01   28.49   <2e-16 ***
NO2          5.819e-02  5.776e-03   10.07   <2e-16 ***
I(NO2^2)    -7.396e-04  7.331e-05  -10.09   <2e-16 ***
PM2.5        5.875e-02  4.209e-03   13.96   <2e-16 ***
I(PM2.5^2)  -8.164e-04  5.545e-05  -14.72   <2e-16 ***
---
Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

(Dispersion parameter for poisson family taken to be 1)

    Null deviance: 3597.6  on 64  degrees of freedom
Residual deviance: 3132.2  on 60  degrees of freedom
AIC: 3598.6

Number of Fisher Scoring iterations: 4

J98

Start:  AIC=541.26
`J:98` ~ NO2 + I(NO2^2) + PM2.5 + I(PM2.5^2)

Call:
glm(formula = `J:98` ~ NO2 + I(NO2^2) + PM2.5 + I(PM2.5^2), family = "poisson", 
    data = m1.vars)

Coefficients:
              Estimate Std. Error z value Pr(>|z|)    
(Intercept)  1.4168071  0.3553203   3.987 6.68e-05 ***
NO2          0.0597128  0.0182508   3.272 0.001069 ** 
I(NO2^2)    -0.0008196  0.0002347  -3.492 0.000479 ***
PM2.5        0.0313581  0.0115102   2.724 0.006442 ** 
I(PM2.5^2)  -0.0003758  0.0001434  -2.621 0.008777 ** 
---
Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

(Dispersion parameter for poisson family taken to be 1)

    Null deviance: 249.11  on 64  degrees of freedom
Residual deviance: 221.72  on 60  degrees of freedom
AIC: 541.26

Number of Fisher Scoring iterations: 4

J40

Start:  AIC=492.92
`J:40` ~ NO2 + PM2.5 + Temp.avg

           Df Deviance    AIC
- PM2.5     1   220.07 491.30
- Temp.avg  1   220.42 491.65
<none>          219.69 492.92
- NO2       1   222.19 493.43

Step:  AIC=491.3
`J:40` ~ NO2 + Temp.avg

           Df Deviance    AIC
- Temp.avg  1   220.66 489.89
<none>          220.07 491.30
- NO2       1   222.51 491.74
+ PM2.5     1   219.69 492.92

Step:  AIC=489.89
`J:40` ~ NO2

           Df Deviance    AIC
- NO2       1   222.56 489.79
<none>          220.66 489.89
+ Temp.avg  1   220.07 491.30
+ PM2.5     1   220.42 491.65

Step:  AIC=489.79
`J:40` ~ 1

           Df Deviance    AIC
<none>          222.56 489.79
+ NO2       1   220.66 489.89
+ PM2.5     1   222.29 491.53
+ Temp.avg  1   222.51 491.74

Call:
glm(formula = `J:40` ~ 1, family = "poisson", data = m1.vars)

Coefficients:
            Estimate Std. Error z value Pr(>|z|)    
(Intercept)  2.39088    0.03753   63.71   <2e-16 ***
---
Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

(Dispersion parameter for poisson family taken to be 1)

    Null deviance: 222.56  on 64  degrees of freedom
Residual deviance: 222.56  on 64  degrees of freedom
AIC: 489.79

Number of Fisher Scoring iterations: 5