Granger Causality Test

This section by utilizing Granger Causality tests which uses statistical tests to determine whether one variable’s time series is useful in forecasting index pricing. Below is a summary of the indexes that passed the test:

Propane FEI

  • NormalButane.MontBelvieu

Butane FEI

  • Propane.FEI
  • Propane.SaudiAramco
  • NormalButane.SaudiAramco

Propane Mont Belvieu

  • NormalButane.MontBelvieu
  • NormalButane.Sonatrach
  • C5_Seller.MontBelvieu

Butane Mont Belvieu

  • Propane.FEI
  • NormalButane.FEI

Propane FEI-Mont Belvieu Differential

NA

Butane FEI-Mont Belvieu Differential

NA

Principal Component Analysis

The Principal Component Analysis is a way of reducing the variance of all the data analyzed. Vectors that most closely related are the most similar with their behavior within the Time-Series. In total, the Biplot below explains ~80% of the TOTAL VARIANCE within the data.

## Warning: `select_()` is deprecated as of dplyr 0.7.0.
## Please use `select()` instead.
## This warning is displayed once every 8 hours.
## Call `lifecycle::last_warnings()` to see where this warning was generated.

Correlation Plots

Below is the correlation plot for Indexes involved. As we can see the differentials of FEI & Mont Belvieu have the lowest correlation to any outside index.