Neural Networks Technologies

Domestic technologies of neural networks predicted a fall in Bernt prices

Some time ago, a self-learning neural network to track the dynamics of world prices on Brent oil has been created in Russia. Now Neural Networks Technologies showed what they are capable of: artificial intelligence was able to make a correct forecast regarding the reduction of the exchange price of the brand. This news was covered in the official press release on the website of the Ministry of Energy of the Russian Federation. Artificial intelligence was designed at the Institute of Energy Strategy with the involvement of scientists from the University of Dubna. The algorithm is based on tracking of recurring pricing trends over a long period of time. The system learns independently, comparing past and current trends, and builds forecasts for the future.

How do neural network technologies work?

Artificial intelligence, created by the Russians, is based on the leading economic pattern – the frequency of rise and decrease in prices for hydrocarbons. In addition, specialists took into account cyclical changes in the form of certain combinations. According to the forecast of the neural network created during the joint work of researchers, until the summer of 2020 we will have a strong and stable decrease in the cost of energy carriers Brent, up to USD 50 per barrel. In the future, until 2026, a calmer behavior of quotations can be expected – until 2021 the tariff will rise slightly, then by 2023 it will fall to the values of USD 60, and again will return to USD 85-90 by December 2024. At least, such data was presented by the neural network.

The Ministry of Energy is convinced that artificial intelligence will prevent a huge number of multi-format and based on non-constant factors of cost forecasts. According to the Ministry of Energy, the same artificial intelligence previously predicted a significant rise in price of Brent oil in 2009, and even the crisis of the hydrocarbons market in 2015, as well as other emerging cases and sharp reversals of quotations on global exchanges. In addition, AI predictions and actual energy prices during the period 2009 – 2018 completely coincided. This suggests that neural networks are really trustworthy, and it sees trends better than any experienced economist.

Some time ago the head of the Ministry of Energy Alexander Novak announced his forecast: for a long time the cost of oil will be about USD 50 per barrel. He stressed that the need for oil products is falling in the world, and the competition between oil producing companies and countries is growing. Because of this, there is a tense situation between countries that are in a hurry to monetize hydrocarbon reserves. According to OPEC, oil production decreased to 32 million barrels per day at the end of 2018. At the same time, in the United States, on the contrary, the turnover is increased to 11.9 million barrels per day, trying to meet the growing demand. In its latest analysis, OPEC estimated the increase in oil consumption to 98.78 million barrels per day. For 2019, the forecast of the organization is 100 million barrels per day. The report also suggests a significant drop in oil production in Norway and Mexico and increasing in Russia, Canada and the United States.