BDI & SCFI prediction (‘22/17W)

 The comparison and test of BDI and SCFI predicted last week and Actual BDI and SCFI last week are compared and tested. Through the process of comparing the prediction error and direction of the predictive value, I will continue to practice the history of the parameter value.

---

1. Last week's forecasts

1-1. BDI, Actual Vs. Predicted

Date	Last week Forecast	Actual BDI	MAPE(%)	Accuracy(%)
‘22/16W(4/18~22)	2,187.904	2,200.65	0.6%	99.4%

 1-2. SCFI, Actual Vs. Predicted

Date	Last week Forecast	Actual SCFI	MAPE(%)	Accuracy(%)
‘22/16W(4/18~22)	4,215.044	4,195.98	0.5%	99.5%

 The prediction accuracy of BDI and SCFI was 99.4% and 99.5%, respectively. The predicted value (2,187.904) of the BDI and the actual value (2,200.65), the predicted value (4,215.044) of the SCFI (4,215.044) and the actual value (4,195.98) have achieved trend and accuracy. In particular, BDI takes great point in predicting directions on the last week.

---

2. BDI forecast

2-1. ARIMA, Time series analysis

 To forecast the BDI, I used weekly data (429) from January 3, 2014 to April 22, 2014 (429),to predict the future BDI for three weeks.

 

Date	Forecast	Trend(WoW)
‘22/17W(4/25~29)	2,246.158	↗
‘22/18W(5/2~6)	2,259.878	→
‘22/19W(5/9~13)	2,257.405	→

 As shown in the table above and in the figure, BDI (4/22: 2,200.65) is expected to rise for the next three weeks for the next three weeks, following the trend after the rise. However, the increase is not large, and the 19th week (5/9 ~ 13) is expected to show a strong joining.

 

Ref) BDI & SCFI prediction (‘22/16W) (tistory.com)

BDI & SCFI prediction (‘22/16W)

2-2. Logistic regression analysis

 Logistic regression analysis was performed with the same data predicted as the Arima model.

 The following statement was higher than 0.566 to 0.566. In other words, the direction of the next-week drive market was predicted in the logistics model. Similar to the ARIMA model, the baseline (0.5) is exceeded, but it is expected to continue to revenge when the number is not large, but when the number is not large.

---

3. SCFI Outlook

 

3-1. ARIMA, Time series analysis

 For SCFI prospects, we used weekly data (627) from October 16, 2009 (627) from April 22, 2009 (627), forecasts for 3 weeks for three weeks.

 

Date	Forecast	Trend(WoW)
‘22/17W(4/25~29)	4,187.446	↘
‘22/18W(5/2~6)	4,177.202	→
‘22/19W(5/9~13)	4,172.082	→

 As shown in the table above and in the figure, SCFI (4/22: 4,195.98) is expected to fall in the 17th to 19th weeks. As the blockade of Shanghai Port, China, will soon be solved, but the supply of ships will increase at the same time, so it is necessary to see how much fare will affect. pls refer to the below.

 

Ref) Increased concerns about supply chain stagnation due to tightening control measures at Chinese ports (tistory.com)

Increased concerns about supply chain stagnation due to tightening control measures at Chinese ports

3-2. Logistic regression analysis

 

 The result of the logistic regression model of SCFI is also the same as the ARIMA value. The predicted value of 0.38 was predicted by declining the fare direction of SCFI next week.

 

Thanks.