Correlation Between Palm Oil Prices and Psi

Part D

Point to note:

1. June 2011 to July 2015 PSI figures were used. These figures were chosen because of the range of monthly palm oil prices that we have extracted from: www.ycharts.com/indicators/palm_oil_price.
2. The PSI figures were calculated based on the average of the lowest and highest PSI of the day. To illustrate, 30 June 2011’s lowest PSI was 20 while the highest PSI was 31, the average of 25.5 was taken as a variable to compare with the price of palm oil. The PSI figures were extracted from: www.nea.gov.sg/anti-pollution-radiation-protection/air-pollution-control/psi/historical -psi-readings.
3. The hotspot locations data available from the NEA website are from 2013 to 2015. The data from October 2013 to February 2014 are missing from the website. The hotspot figures were extracted from: www.nea.gov.sg/anti-pollution-radiation-protection/air-pollution-control/psi/hotspot-and-satellite-images.

A quick scan shows that the PSI stats in September 2014 and 2015 are generally higher than the rest of the months. There is also a general decreasing palm oil prices trend observed since 2011 to 2015.

A statistical test of the PSI and palm oil prices variables was conducted. Using StatTools, the R2 gotten for the palm oil price is 0.1160. This means that the PSI of Singapore is not adequate enough to explain the variation of palm oil prices. There is still 88.4% of the variation left unexplained.

Following are some of the reasons why there is no/weak correlation between PSI and palm oil prices:

1. There are other stronger variables that affect the palm oil prices. Firstly, the supply and demand of palm oil. Less demand will push the price of palm oil down since it will lead to a supply surplus of palm oil.
2. The weather patterns such as seasonal rains and drought could also affect the growth of oil palm plantation leading to lower palm oil production.
3. The PSI in Singapore might not be a good representation of the ground condition. Since the amount of haze in Singapore is a result of the direction of the wind. A low PSI in Singapore might not mean that there are no hotspots; it could just be a result of the wind direction. This point is further supported by the weak correlation that was observed between the PSI (average) and the number of hotspots observed from the Google Earth images from NEA.
4. The price of competing vegetable oils such as soybean oil and corn oil can also affect the price of the palm oil. For example, a lower production of soybean oil will increase in the demand for palm oil, thus increasing palm oil prices.

Therefore, we want to find another predictor for the palm oil prices. We will be using soybean oil (the next largest edible oil consumed) as another predictor of palm oil prices. CME Group has also published a report to show the positive price relationships between the two oils.

A statistical test of soybean oil and palm oil prices variables was conducted. Using StatTools, the calculated R2 was 0.7161. This means that there is a strong positive relationship between soybean and palm oil prices. The equation will be: predicted palm oil price = 239.08 + 0.6602 (soybean oil price).