BDI & SCFI prediction (‘22/20W)

In the 20th week, BDI and SCFI predicted values and actual values are compared and tested.

---

1.     Last Week’s forecasts

1-1.        BDI, Actual Vs. Predicted

Date	Last week Forecast	Actual BDI	MAPE(%)	Accuracy(%)
‘22/19W(5/9~13)	2,623.855	3,008.60	12.8%	87.2%

 

1-2.        SCFI, Actual Vs. Predicted

Date	Last week Forecast	Actual SCFI	MAPE(%)	Accuracy(%)
‘22/19W(5/9~13)	4,155.752	4,147.83	0.2%	99.8%

 The prediction accuracy of BDI and SCFI was 87.2% and 99.8%, respectively. The predicted values (2,623.855) and actual values (3,008.60) of BDI, and predicted values (4,155.752) and actual values (4,147.83) of SCFI all achieved Trend and Accuracy. The average BDI MAPE value for the last seven predictions is 4.5% and the prediction rate is 95.5%, and the SCFI average MAPE value is 0.7% and the prediction rate is 99.3%. However, the accuracy was slightly lowered because the increase in the actual value compared to the BDI predicted value was large.

---

2.     BDI forecast

2-1. ARIMA, Time series analysis

 For the BDI index forecast, I used weekly data (432 pieces) from January 3, 2014 to May 13, 2022 to predict the freight for three weeks in the future.

Date	Forecast	Trend(WoW)
‘22/20W(5/16~20)	3,203.596	↑
‘22/21W(5/23~27)	3,213.826	↑
‘22/22W(5/30~6.3)	3,157.828	↘

 

 As shown in the table and figure above, the BDI (as of May 13th: 3,008.60) rose 17.3% from the previous week (as of May 6th: 2,564.75) and reached a new high in 2022. Considering the fact that the Cape charter rate and Panamax charter rate are still similar, we believe there is additional upside potential for Cape. The ARIMA model value also predicted that the BDI would rise for the next 1-2 weeks.

Ref) BDI & SCFI prediction (‘22/19W) (tistory.com)

BDI & SCFI prediction (‘22/19W)

 2-2. Logistic regression analysis

Logistic regression analysis was performed with the same data predicted by the ARIMA model previously.

 The expected trend for next week was 0.674, which was higher than the baseline 0.5. However, considering that it has decreased slightly compared to the predicted value of the 19th week (0.8), the increase is expected to slow down.

---

 3.     SCFI Outlook

3-1. ARIMA, Time series analysis

 For the SCFI index forecast, I used weekly data (630 pieces) from October 16, 2009 to May 13, 2022 to predict the future freight for three weeks.

Date	Forecast	Trend(WoW)
‘22/20W(5/16~20)	4,137.046	↘
‘22/21W(5/23~27)	4,126.881	↘
‘22/22W(5/30~6.3)	4,118.121	↘

 As shown in the table and figure above, the SCFI (as of 5/13: 4,147.83) is expected to show a continuous decline following last week's forecast. Despite the 17 consecutive weeks of decline, the end of the decline is visible little by little, considering that the West Coast route to the Americas has turned upward. The detailed news related to the rise by route will be dealt with in the ‘Container Shipping Market ('22/20W)’.

 

3-2. Logistic regression analysis

 The results of the logistic regression analysis model of SCFI are also the same as the ARIMA values. With the predicted value of 0.388, the next week's SCFI freight direction was predicted to be down.

 

3-3. SVR model

 

Ref) SCFI Prediction Using SVR Model (tistory.com)

SCFI Prediction Using SVR Model

 For the SCFI index forecast, I used weekly data (71 pieces) from January 01, 2021 to May 13, 2022 to predict the future freight for three weeks. Since I used the latest data from 2021, the data sample itself is small, but the current trend can be confirmed a little more. As a result of the model, it is predicted that the value will continue to decline as follows.

Date	Forecast	Trend(WoW)
‘22/20W(5/16~20)	3,994.327	↓
‘22/21W(5/23~27)	3,951.649	↓
‘22/22W(5/30~6.3)	3,909.825	↓

 

Thanks.