輸入values

DATE <- c("324", "325", "326", "327", "328", "329", "330", "331", "401",
          "402", "403", "404", "405", "406", "407", "408", "409", "410", "411","412")
Domestic <- c(15, 14, 21, 83, 34, 33, 56, 87, 104, 160, 183, 133, 216, 281, 382, 384, 442, 431, 439, 511)
Oversea <- c(124, 122, 82, 120, 93, 63, 107, 152, 132, 244, 97, 142, 65, 78, 149, 123, 136, 144, 191, 112)

create dataframe

covid <- data.frame(DATE, Domestic, Oversea)
scatterplot(Domestic ~ Oversea, 
            data = covid, 
            smooth = F)

#灰色範圍越小越準確
ggplot(aes(x = Oversea, y = Domestic), data = covid ) +
  geom_point() +
  geom_smooth(method = lm) +
  theme_bw()
## `geom_smooth()` using formula 'y ~ x'

#連續變項
cvmod <- lm(Domestic ~ Oversea, data = covid)
summary(cvmod)
## 
## Call:
## lm(formula = Domestic ~ Oversea, data = covid)
## 
## Residuals:
##     Min      1Q  Median      3Q     Max 
## -185.66 -134.49  -95.09  158.33  322.93 
## 
## Coefficients:
##             Estimate Std. Error t value Pr(>|t|)
## (Intercept)  70.5145   119.6416   0.589    0.563
## Oversea       1.0496     0.9163   1.145    0.267
## 
## Residual standard error: 170.1 on 18 degrees of freedom
## Multiple R-squared:  0.06794,    Adjusted R-squared:  0.01616 
## F-statistic: 1.312 on 1 and 18 DF,  p-value: 0.267

p-value > 0.05，無法虛無假設，且t-value與0距離1.145，不具有足夠解釋力。 R-squared = 0.01616，Domestic中只有0.01616的變異量可以被oversea預測，表示有其他factor能更有效解釋。

因為多數本土案例集中於北北基地區，下雨時民眾比較傾向待在家，而無雨時則會出門活動，故加入變項:臺北氣象站逐日雨量

rain <- c(0,0.5,14.5,42,75,11.5,0,0,0,23,10.5,0,0,0,0,0,0,0,0,0)
cvrain <- data.frame(DATE, Domestic, rain)
ggplot(aes(x = rain, y = Domestic), data = cvrain ) +
  geom_point() +
  geom_smooth(method = lm) +
  theme_bw()
## `geom_smooth()` using formula 'y ~ x'

cvRain <- lm(Domestic ~ rain, data = cvrain)
summary(cvRain)
## 
## Call:
## lm(formula = Domestic ~ rain, data = cvrain)
## 
## Residuals:
##     Min      1Q  Median      3Q     Max 
## -215.56 -147.28   -8.25  151.94  280.44 
## 
## Coefficients:
##             Estimate Std. Error t value Pr(>|t|)    
## (Intercept)  230.560     40.496   5.693 2.12e-05 ***
## rain          -3.402      1.979  -1.719    0.103    
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 163.3 on 18 degrees of freedom
## Multiple R-squared:  0.141,  Adjusted R-squared:  0.09327 
## F-statistic: 2.954 on 1 and 18 DF,  p-value: 0.1028

p-value > 0.05，無法拒絕虛無假設，且t-value與0距離1.719，獨變項沒有足夠的解釋力。 R-squared = 0.09327，Domestic中只有9.327%的變異量可以被每日雨量預測，表示有其他factor能更有效解釋。

因為多數本土案例集中於北北基地區，加入變項:每日紫外線量

UV <- c(7,11,7,5,3,7,8,7,3,3,3,10,11,11,8,11,12,12,13,12)
cvUV <- data.frame(DATE, Domestic, UV)
ggplot(aes(x = UV, y = Domestic), data = cvUV ) +
  geom_point() +
  geom_smooth(method = lm) +
  theme_bw()
## `geom_smooth()` using formula 'y ~ x'

cvuvlm <- lm(Domestic ~ UV, data = cvUV)
summary(cvuvlm)
## 
## Call:
## lm(formula = Domestic ~ UV, data = cvUV)
## 
## Residuals:
##      Min       1Q   Median       3Q      Max 
## -274.357 -125.415   -5.276  113.998  191.248 
## 
## Coefficients:
##             Estimate Std. Error t value Pr(>|t|)   
## (Intercept)  -56.991     80.747  -0.706  0.48935   
## UV            31.395      9.114   3.445  0.00289 **
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 136.8 on 18 degrees of freedom
## Multiple R-squared:  0.3973, Adjusted R-squared:  0.3638 
## F-statistic: 11.87 on 1 and 18 DF,  p-value: 0.002889

p-value < 0.05，拒絕虛無假設，且t-value與0距離3.445，獨變項有解釋力。 R-squared = 0.3638，Domestic中有36.38%的變異量可以被每日紫外線量預測。

新增變項:每日台灣新冠疫苗施打兩劑占總人口比率

VX <- c(76.39,76.43,76.43,76.46,76.46,76.52,76.56,76.60,76.65,76.65,76.65,76.65,76.65,76.73,76.78,76.84,76.91,76.98,77.04,77.06)
cvVX <- data.frame(DATE, Domestic, VX)
ggplot(aes(x = VX, y = Domestic), data = cvVX ) +
  geom_point() +
  geom_smooth(method = lm) +
  theme_bw()
## `geom_smooth()` using formula 'y ~ x'

cvVXlm <- lm(Domestic ~ VX, data = cvVX)
summary(cvVXlm)
## 
## Call:
## lm(formula = Domestic ~ VX, data = cvVX)
## 
## Residuals:
##     Min      1Q  Median      3Q     Max 
## -78.881 -47.020   1.775  35.612  95.304 
## 
## Coefficients:
##              Estimate Std. Error t value Pr(>|t|)    
## (Intercept) -61027.99    4102.38  -14.88 1.48e-11 ***
## VX             798.58      53.51   14.93 1.40e-11 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 48.17 on 18 degrees of freedom
## Multiple R-squared:  0.9252, Adjusted R-squared:  0.9211 
## F-statistic: 222.8 on 1 and 18 DF,  p-value: 1.402e-11

p-value < 0.05，拒絕虛無假設，且t-value與0距離14，獨變項有解釋力。 R-squared = 0.9211，Domestic中有92.11%的變異量可以被每日台灣新冠疫苗施打兩劑占總人口比率解釋。