412

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, 551)

Oversea <- c(124, 122, 82, 120, 93, 63, 107, 152, 132, 244, 97, 142, 65, 78, 149, 123, 136, 144, 191, 112)

Hsinchu <- c(1, 1, 2, 1, 1, 2, 1, 2, 1, 2, 2, 7, 8, 4, 5, 3, 6, 10, 10, 8)

#將Domestic、Hsinchu和Oversea存入cvdta
cvdta <- data.frame(DATE, Domestic, Oversea, Hsinchu)

#將DATE轉為類別變項
cvdta$DATE <- as.factor(cvdta$DATE)

#繪製Domestic和Oversea的相關圖
scatterplot(Oversea ~ Domestic, 
            data = cvdta, 
            smooth = F)

#繪製Domestic和Oversea的相關圖
ggplot(aes(x = Domestic, y = Oversea), data = cvdta) +
  geom_point() +
  geom_smooth(method = lm, se = T) +
  theme_bw()

## `geom_smooth()` using formula 'y ~ x'

從2個圖中可以看見國外病例和本土病例的增加趨勢呈現正相關，但整體落點是分散的

#散佈圖 加上模型預測區域 
cvLM <- lm(Oversea ~ Domestic, data = cvdta)

summary(cvLM)

## 
## Call:
## lm(formula = Oversea ~ Domestic, data = cvdta)
## 
## Residuals:
##     Min      1Q  Median      3Q     Max 
## -59.627 -26.893   0.536  14.216 122.790 
## 
## Coefficients:
##              Estimate Std. Error t value Pr(>|t|)    
## (Intercept) 111.44894   14.67281   7.596 5.09e-07 ***
## Domestic      0.06101    0.05536   1.102    0.285    
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 42.35 on 18 degrees of freedom
## Multiple R-squared:  0.0632, Adjusted R-squared:  0.01115 
## F-statistic: 1.214 on 1 and 18 DF,  p-value: 0.285

虛無假設：Oversea變異數=Domestic變異數

對立假設：Oversea變異數≠Domestic變異數

R-squared數值為0.01115，預測解釋力低，且p-value未達顯著水準，因此可以推論Domestic的數值無法有效預測Oversea。

#繪製Domestic和Hsinchu的相關圖
scatterplot(Hsinchu ~ Domestic, 
            data = cvdta, 
            smooth = F)

#繪製Domestic和Hsinchu的相關圖
ggplot(aes(x = Domestic, y = Hsinchu), data = cvdta) +
  geom_point() +
  geom_smooth(method = lm, se = T) +
  theme_bw()

## `geom_smooth()` using formula 'y ~ x'

從2個圖中可以看見新竹病例和本土病例的增加趨勢呈現正相關，且整體收斂程度比國外病例好，可以預期本土病例與新竹病例之間的相關比本土病例對國外更高。

#散佈圖 加上模型預測區域 
cvLM2 <- lm(Hsinchu ~ Domestic, data = cvdta)

summary(cvLM2)

## 
## Call:
## lm(formula = Hsinchu ~ Domestic, data = cvdta)
## 
## Residuals:
##     Min      1Q  Median      3Q     Max 
## -3.4228 -1.2456 -0.6188  0.5938  4.1342 
## 
## Coefficients:
##             Estimate Std. Error t value Pr(>|t|)    
## (Intercept)  0.98106    0.70760   1.386    0.183    
## Domestic     0.01417    0.00267   5.308 4.79e-05 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 2.042 on 18 degrees of freedom
## Multiple R-squared:  0.6101, Adjusted R-squared:  0.5885 
## F-statistic: 28.17 on 1 and 18 DF,  p-value: 4.794e-05

虛無假設：Hsinchu變異數=Domestic變異數

對立假設：Hsinchu變異數≠Domestic變異數

R-squared數值為0.5885，預測解釋力比對國外病例高，且p-value達顯著水準，因此可以推論Domestic的數值或許更能有效預測Hsinchu的病例增長趨勢。