## Warning: Removed 3242 rows containing missing values (`geom_line()`).
## # A tsibble: 262 x 10 [1Y]
## # Key: Country [262]
## Country Code Year GDP Growth CPI Imports Exports Population gperc
## <fct> <fct> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
## 1 Luxembourg LUX 2017 6.24e10 2.30 111. 194. 230. 599449 1.04e5
## 2 Macao SAR… MAC 2017 5.04e10 9.10 136. 32.0 79.4 622567 8.09e4
## 3 Switzerla… CHE 2017 6.79e11 1.09 98.3 53.9 65.0 8466017 8.02e4
## 4 Norway NOR 2017 3.99e11 1.92 115. 33.1 35.5 5282223 7.55e4
## 5 Iceland ISL 2017 2.39e10 3.64 122. 42.8 47.0 341284 7.01e4
## 6 Ireland IRL 2017 3.34e11 7.80 105. 87.9 120. 4813608 6.93e4
## 7 Qatar QAT 2017 1.67e11 1.58 116. 37.3 51.0 2639211 6.32e4
## 8 United St… USA 2017 1.94e13 2.27 112. NA NA 325719178 5.95e4
## 9 North Ame… NAC 2017 2.10e13 2.35 NA NA NA 362492702 5.81e4
## 10 Singapore SGP 2017 3.24e11 3.62 113. 149. 173. 5612253 5.77e4
## # ℹ 252 more rows
Luxembourg has the highest GDP per capita in 2017. As we can see in the plot above, Luxembourg began increasing exponentially over time at a fairly consistent rate. From 2008 to 2017, the GDP per capita seemed to fluctuate throughout the remaining years.
Seeing as this is difficult to interpret, it would be best to change the time interval from 30 minutes to each day over the course of 3 years. Additionally, the average of each demand within each day must be calculated to receive the average demand. The plot below is easier to interpret and consists of a similar scale the initial demand contained.
It is difficult to interpret the trend that occurs between 1956 and 1970. To enhance this, Box-Cox transformation should be applied. With guerrero, λ was selected at 0.11, which allows the plot to be more interpretable and consistent.
Box-Cox transformation will not be helpful due to the seasonal variation being fairly consistent across the series.
For this specific seed, the guerrero feature selected λ = 0.08 to make the variance more stable. As we can see, there is less of a spread in the data in the later seasons.
## Warning: Removed 24 rows containing missing values (`geom_line()`).
## Warning: Removed 24 rows containing missing values (`geom_line()`).
The trend-cycle increases over the 5 years. The seasonal fluctuations show highs in quarter 3 and lows in quarter 1 consistently each year.
## Warning: Removed 2 rows containing missing values (`geom_line()`).
Yes, as the trend is increasing over each quarter and the seasonal lows are Q1 and highs are Q3.
## Warning: Removed 4 rows containing missing values (`geom_line()`).
## Warning: Removed 2 rows containing missing values (`geom_line()`).
## Warning: Removed 2 rows containing missing values (`geom_line()`).
## Warning: Removed 2 rows containing missing values (`geom_line()`).
Wherever an outlier is placed, you notice the trend peaks and return to its gradual increase. The seasonal lows are in Q1 no matter where the outlier is placed. However, the seasonal highs are in Q3 for outliers in the front and end, but Q4 when the outlier is in the middle.
The seasonal plot shows a mirror in the results, where the downward spike becomes the peak spike as the months increase. The trend remains to increase over time with slight “hiccups” throughout the months.
There is a seasonality that consists of similar values to the remainder. There is, however, a slight dip around 1991 due to the recession at the time. The trend increases at a similar slope of the value plot.
In the remainder component plot, you can see the recession of 1991/1992 due to the dip close to -400.