library(ggplot2); library(readxl); library(ggrepel); library(tidyverse); library(gridExtra); library(ggthemes)
## Warning: package 'ggplot2' was built under R version 4.0.2
## Warning: package 'tidyverse' was built under R version 4.0.2
## ── Attaching packages ─────────────────────────────────────── tidyverse 1.3.1 ──
## ✓ tibble 3.1.2 ✓ dplyr 1.0.7
## ✓ tidyr 1.1.3 ✓ stringr 1.4.0
## ✓ readr 1.4.0 ✓ forcats 0.5.1
## ✓ purrr 0.3.4
## Warning: package 'tibble' was built under R version 4.0.2
## Warning: package 'tidyr' was built under R version 4.0.2
## Warning: package 'readr' was built under R version 4.0.2
## Warning: package 'dplyr' was built under R version 4.0.2
## Warning: package 'stringr' was built under R version 4.0.2
## Warning: package 'forcats' was built under R version 4.0.2
## ── Conflicts ────────────────────────────────────────── tidyverse_conflicts() ──
## x dplyr::filter() masks stats::filter()
## x dplyr::lag() masks stats::lag()
##
## Attaching package: 'gridExtra'
## The following object is masked from 'package:dplyr':
##
## combine
df = read_excel("~/Dropbox/Bao chi/Khoa hoc & Y te/Mortality analysis VN/CFR in VN 28-9-2021.xlsx")
df1 = subset(df, log(Cases)>4.9 & Deaths>0)
Số ca tử vong và số ca nhiễm
ggplot(data=df, aes(x=log(Cases), y=log(Deaths), label=Pr, col=Pr)) + geom_point() + geom_text_repel() + theme(legend.position="none") + labs(x="Số ca (log)", y="Số tử vong (log)")
Mật độ dân số và CFR
ggplot(data=df, aes(x=log(Density), y=CFR, label=Pr, col=Pr)) + geom_point() + geom_text_repel() + theme(legend.position="none") + labs(x="Mật độ dân số (log)", y="Tỉ lệ tử vong (CFR)")
ggplot(data=df, aes(x=log(Density), y=log(Cases), label=Pr, col=Pr)) + geom_point() + geom_text_repel() + theme(legend.position="none") + labs(x="Mật độ dân số (log)", y="Số ca (log)")
Số ca tử vong và số ca nhiễm
ggplot(data=df1, aes(x=log(Cases), y=log(Deaths), label=Pr)) + geom_point() + geom_smooth(method="lm") + geom_text_repel(aes(col=Pr)) + theme(legend.position="none") + labs(x="Số ca (log)", y="Số tử vong (log)")
## `geom_smooth()` using formula 'y ~ x'
Poisson regression
p1 = glm(Deaths ~ Prov + offset(log(Cases)), family=poisson(link="log"), data=df1)
summary(p1)
##
## Call:
## glm(formula = Deaths ~ Prov + offset(log(Cases)), family = poisson(link = "log"),
## data = df1)
##
## Deviance Residuals:
## [1] 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
## [26] 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -5.18448 0.20000 -25.922 < 2e-16 ***
## ProvBà Rịa - Vũng Tàu 0.64978 0.24847 2.615 0.00892 **
## ProvBắc Giang -0.93091 0.34194 -2.722 0.00648 **
## ProvBạc Liêu -0.14340 0.73485 -0.195 0.84528
## ProvBắc Ninh 0.34187 0.32660 1.047 0.29522
## ProvBến Tre 1.51154 0.24578 6.150 7.75e-10 ***
## ProvBình Định 0.54813 0.35119 1.561 0.11857
## ProvBình Dương 0.49765 0.20133 2.472 0.01344 *
## ProvBình Phước -0.07227 0.42762 -0.169 0.86578
## ProvBình Thuận 1.19645 0.23925 5.001 5.71e-07 ***
## ProvCà Mau 0.70998 0.53852 1.318 0.18737
## ProvCần Thơ 1.05156 0.22635 4.646 3.39e-06 ***
## ProvĐà Nẵng 1.19309 0.22361 5.336 9.52e-08 ***
## ProvĐắk Lắk -0.27511 0.42762 -0.643 0.52000
## ProvĐắk Nông -1.37938 1.01980 -1.353 0.17618
## ProvĐồng Nai 0.39185 0.20641 1.898 0.05764 .
## ProvĐồng Tháp 1.14203 0.21667 5.271 1.36e-07 ***
## ProvGia Lai -0.47326 0.73485 -0.644 0.51956
## ProvHà Nội 0.71362 0.24415 2.923 0.00347 **
## ProvHà Tĩnh 0.65403 0.48990 1.335 0.18187
## ProvHải Dương -1.67933 1.01980 -1.647 0.09961 .
## ProvHậu Giang -0.39147 0.73485 -0.533 0.59422
## ProvHưng Yên -0.70993 1.01980 -0.696 0.48634
## ProvKhánh Hòa 0.70688 0.22635 3.123 0.00179 **
## ProvKiên Giang 0.43107 0.24754 1.741 0.08161 .
## ProvLạng Sơn -0.18616 1.01980 -0.183 0.85515
## ProvLong An 0.80917 0.20611 3.926 8.64e-05 ***
## ProvNghệ An 0.44773 0.32016 1.398 0.16197
## ProvNinh Thuận 0.43846 0.40620 1.079 0.28041
## ProvPhú Yên 0.68271 0.26346 2.591 0.00956 **
## ProvQuảng Bình -1.47482 0.73485 -2.007 0.04475 *
## ProvQuảng Nam 0.57308 0.40620 1.411 0.15830
## ProvQuảng Trị -0.15786 1.01980 -0.155 0.87698
## ProvSóc Trăng 1.19457 0.30000 3.982 6.84e-05 ***
## ProvTây Ninh 1.11384 0.21799 5.110 3.23e-07 ***
## ProvThanh Hóa 0.37433 0.53852 0.695 0.48699
## ProvThừa Thiên Huế 0.87185 0.36181 2.410 0.01597 *
## ProvTiền Giang 1.48069 0.20737 7.140 9.31e-13 ***
## ProvTP. Hồ Chí Minh 1.92827 0.20017 9.633 < 2e-16 ***
## ProvTrà Vinh 0.73150 0.31436 2.327 0.01997 *
## ProvVĩnh Long 1.25631 0.25151 4.995 5.88e-07 ***
## ProvVĩnh Phúc 0.73791 0.61101 1.208 0.22717
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for poisson family taken to be 1)
##
## Null deviance: 7.0311e+03 on 41 degrees of freedom
## Residual deviance: -4.4675e-13 on 0 degrees of freedom
## AIC: 286.17
##
## Number of Fisher Scoring iterations: 3