Page 323 - Data Science Algorithms in a Week
P. 323

304             Mayra Bornacelli, Edgar Gutierrez and John Pastrana

                            UK created a tax (Carbon Floor) above the price of CO2 – This tax artificially
                              benefits the generation of less CO2 emitting energy (renewable and natural gas)
                              over all existing technologies with a direct impact on energy costs to the end user
                            Development of the tracking method to extract the shale gas profitably - The
                              cost  effective  gas  produced  with  this  method  displaced  part  of  the  coal  in the
                              USA. The coal that was not consumed locally then began to be exported, which
                              increased the world supply and therefore reduced prices.

                          The problem that we try to solve is summarized in three aspects:

                          1.  Markets such as the oil and coal are oligopolies, which means, the fluctuations of
                              their prices is determined by variables that shape their demand and offer in the
                              market.
                          2.  Over time, analysts have identified some of these variables (and even introduced
                              new ones).  However,  the  relationship  between  the  variables  and  their  order  of
                              importance is not clear yet. This type of study is relevant to find patterns with
                              respect to the price and not analyzing independent events.
                          3.  Each of the variables that have shaped the coal price, have generated their own
                              strength  (positive  or  negative)  in  the  price,  and  about  these  events  the
                              stakeholders have historically reacted.

                          The  objective  of  this  research is to  determine the  most influential  variables in  the
                       price of thermal coal by using the Delphi methodology and the subsequent evaluation of
                       the results with AI techniques such as neural networks and regression trees.


                                                     METHODOLOGY

                          This project proposes an analytical framework that allows managers to analyze prices
                       in the thermal coal industry. Figure 1 shows the general research framework. From the
                       data acquisition, data process, the use of models, and their outputs. With this framework
                       analyzers  may  have  a  tool  to  deal  with  volume  of  data  and  diversity,  handle  the
                       imprecisions and provide robust solutions for price prediction.
                          This process determines challenges and opportunities that a company could face from
                       the data gathering until their analysis and use to create value and optimize their business
                       model.
   318   319   320   321   322   323   324   325   326   327   328