terramodel linetypes not plotting properly

In other words, we suspect that the two time series are cointegrated. Logically, if HSFO prices stabilise, then so should Brent crude oil prices, and vice versa. In the context of our analysis, we believe that there exists a long-run equilibrium between HSFO and Brent crude oil prices, because they are both driven by similar factors, such as the general demand for petroleum products and the level of oil production. In this case, we say that the two series are cointegrated. 3 The exception is when the two time series are involved in some long-run equilibrium, arising from common factors that are able to explain changes in both series, such that when one variable converges to some value the other would do so as well. However, when dealing with time series that show trends, one has to be careful to avoid spurious regressions - “nonsense” regressions that imply correlations where there are none. Visually, both HSFO and Brent prices seem to trend together, implying possible correlation between the two. # read in dataset for HSFO and Brent crude oil prices price_data % mutate ( date2 = dmy ( date )) %>% gather ( brent : HSFO, key = type, value = price ) %>% ggplot ( aes ( x = date2, y = price )) + geom_line ( aes ( linetype = type, color = type ), size = 0.5 ) + ggtitle ( 'Fig.1 Historical HSFO and Brent crude oil prices ( - )' ) + labs ( x = 'Time', y = 'Price (USD per bbl)' ) + scale_color_discrete ( name = 'Series', labels = c ( 'Brent', 'HSFO' )) + scale_linetype_discrete ( name = 'Series', labels = c ( 'Brent', 'HSFO' )) + scale_x_date ( date_breaks = '6 months', date_labels = '%d %b %y' ) + scale_y_continuous ( breaks = seq ( 20, 130, 20 )) + theme ( axis.title = element_text ( size = 14 ), axis.text = element_text ( size = 12 ), = element_text ( angle = 45, hjust = 1 ), legend.title = element_text ( size = 14 ), legend.text = element_text ( size = 12 ), title = element_text ( size = 16 ))

2 Figure 1 below shows a plot of the data. We obtained data on HSFO and Brent crude oil prices from to, measured daily, 1 from Platts. Without diving into the complex economic and social factors that govern oil prices, we attempt to investigate the statistical relationship between historical HSFO and Brent crude oil prices using time series econometrics. Part 1: Modelling the historical relationship between HSFO and Brent crude oil prices All statistical analysis was done in R, Version 3.4.1. This post comes in two parts that respectively address (1) and (2) above.

use this historical relationship and EIA’s projection of Brent crude oil prices to project future HSFO prices.

statistically model the historical relationship between high sulphur fuel oil (HSFO) and Brent crude oil prices, and.This made for a pretty fun econometric exercise that required us to: For a certain project at work, we needed a more informed projection of future high sulphur fuel oil price, instead of assuming a flat average all the way until 2050.

Terramodel linetypes not plotting properly