KR20170036385A - Estimating method and apparatus for crack of Naphtha - Google Patents
Estimating method and apparatus for crack of Naphtha Download PDFInfo
- Publication number
- KR20170036385A KR20170036385A KR1020150135481A KR20150135481A KR20170036385A KR 20170036385 A KR20170036385 A KR 20170036385A KR 1020150135481 A KR1020150135481 A KR 1020150135481A KR 20150135481 A KR20150135481 A KR 20150135481A KR 20170036385 A KR20170036385 A KR 20170036385A
- Authority
- KR
- South Korea
- Prior art keywords
- crack
- naphtha
- cracks
- linear relationship
- price
- Prior art date
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/06—Buying, selling or leasing transactions
- G06Q30/0601—Electronic shopping [e-shopping]
Landscapes
- Business, Economics & Management (AREA)
- Engineering & Computer Science (AREA)
- Strategic Management (AREA)
- Economics (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Finance (AREA)
- Accounting & Taxation (AREA)
- Marketing (AREA)
- Theoretical Computer Science (AREA)
- Human Resources & Organizations (AREA)
- General Business, Economics & Management (AREA)
- Development Economics (AREA)
- Tourism & Hospitality (AREA)
- Quality & Reliability (AREA)
- Operations Research (AREA)
- Entrepreneurship & Innovation (AREA)
- Game Theory and Decision Science (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
According to one aspect of the present invention, there is provided a method and apparatus for predicting a crack in naphtha cracks, (A) analyzing a plurality of independent variables including a price difference between gasoline (Mogas) and naphtha into a plurality of principal components through a principal component analysis technique using an eigen value and an accumulated value; (B) constructing a regression model to regress the correlation between the plurality of principal components and the predicted crack; And (c) calculating a principal component and a crack predicted value through a regression model, respectively.
Description
The present invention relates to a method and apparatus for predicting naphtha cracks.
In general, naphtha (naphtha) is the lightest and volatile component of liquid hydrocarbons in petroleum. It is the product of the refinery and the main raw material of the petrochemical process, divided into various types according to the properties and boiling point.
At this time, the price of naphtha is closely related to the price of crude oil, and it fluctuates in real time, and it is highly estimated by month (Half month).
On the other hand, the naphtha is made through forward trading, and the price is divided into trading regions by MOPJ (Mean of Platt's CFR JAPAN), MOPS (Mean of FOB Singapore) and CIF NWE (CIF North West Europe). Here, MOPJ is the price of naphtha based on the arrival rate in Japan (based on CFR CHIVA terms), and the daily price is calculated and announced by the petrochemical magazine Platts. In addition, MOPJ is the standard index of naphtha price of Asia due to high frequency and reliability of quotation, and it is the standard of other naphtha prices such as FOB (free on board) ARAB GULF and FOB SINGAPORE price.
Normally, the naphtha is calculated as 25,000 MT to 30,000 MT (Metric Tonne) as 1 CARGO, and the arrival of the cargo within the half-month period (15-day interval, hereinafter referred to as 'H') becomes the transaction standard. For example, if you arrive between December 1st and 15th, it is called 1H DEC (1st Half December) CARGO. If you arrive between December 16th and 31st, it is called 2H DEC (2nd Half December) CARGO. In the real market, the month-to-month is the most traded two months after the present, and MOPJ is usually calculated based on this. To explain the specific pricing method, we calculate the average of the 2H prices after 2 months (after 'Today' +45 days, ie after + 3H), including the half-month period in which 'Today' do.
For example, in the case of the naphtha price of November 25, the half-month interval corresponding to November 25 is 2H NOV, so + 3H is 1H JAN, so the average price of 2H JAN and 1H FEB It becomes the naphtha price of 25th month. For example, if 2HJAN water is traded at $ 180-182 per MT, and 1H FEB water is traded at $ 178-180, MOPJ is estimated at $ 180 per ton, which is an average of $ 181 and $ 179.
The purchase period is usually two months (45 to 60 days) before the arrival half-month (Half-month). Pricing is determined by the price difference between the month of purchase at the time of purchase and the MOPJ of the business day of one or two weeks Spread (premium) is added or subtracted. That is, at the time of purchase, it is a general type of purchasing to purchase only by determining the price fixing interval and the premium level without confirming the price.
Therefore, it is very important to predict the premium when purchasing naphtha.
On the other hand, cracks represent the price difference between Brent oil and naphtha for each month, which is a margin of naphtha to crude oil in terms of physical meaning or economy. Cracks are often used as a buying strategy in naphtha trading, and trading with Brent cracks is an essential form of trading in naphtha trading. Securing a range of crude trading is very important in trading activities.
An object of the present invention is to provide a method and apparatus for predicting a naphtha crack that can predict a naphtha crack utilized as an index of naphtha trading.
According to an aspect of the present invention, there is provided a method for predicting a crack, which is a price difference between Brent oil and Naptha, (A) analyzing a plurality of independent variables including a price difference of a naphtha into a plurality of principal components through a principal component analysis technique using an eigen value and an accumulated value; (B) constructing a regression model to regress the correlation between the plurality of principal components and the predicted crack; And (c) calculating a principal component and a crack predicted value through a regression model, respectively.
According to another aspect of the present invention, there is provided a device for predicting a crack, which is a price difference between Brent oil and Naptha, an input unit for inputting a plurality of independent variables including a spread; And the input independent variables are analyzed by principal component analysis using eigenvalue and cumulative value, and a regression model is constructed to regress the correlation between plural principal components and predicted cracks. There is provided a naphtha crack prediction apparatus including a control unit for calculating a main component and a crack predicted value, respectively.
As described above, according to the method and apparatus for predicting naphtha cracks according to an embodiment of the present invention, principal components are derived from a plurality of independent variables related to naphtha cracks through a principal component analysis technique, Can be predicted.
1 is a graph for explaining naphtha trading using a crack.
2 is a flowchart showing a method of predicting a naphtha crack according to an embodiment of the present invention.
3 is a graph showing the relationship between the first main component and cracks.
4 is a graph showing the relationship between the second main component and cracks.
5 is a graph showing the relationship between the third main component and cracks.
Hereinafter, a method and apparatus for predicting naphtha cracks according to an embodiment of the present invention will be described in detail with reference to the accompanying drawings.
In addition, the same or corresponding reference numerals are given to the same or corresponding reference numerals regardless of the reference numerals, and redundant description thereof will be omitted. For convenience of explanation, the size and shape of each constituent member shown in the drawings are exaggerated or reduced .
Unlike financial assets, crude oil is mainly used for consumption, so the risk from supply disruption is greater than other assets. Therefore, the convenience yield of the holding of the spot increases. The benefit is the benefit that arises from holding the substance directly rather than the futures contract, reflecting market expectations for future asset usability. If short-term or long-term supply shortages are likely to occur, real users are willing to hold assets (naphtha) at a certain cost. The futures price is determined by the spot price, the interest rate, the storage cost and the convenience profit, and the phenomenon that the spot price is relatively expensive (the futures price <spot price) is called the backwarding. The period structure of oil prices is generally backward, in which the spot price is higher than the futures price, and premium is generated at the time of purchase of naphtha.
Conversely, if the balance of supply and demand is stable, or if the supply is expected to be abundant, there is no reason for the buyer to pay the extra cost and hold the raw materials in a hurry. Therefore, as the profit of convenience is lowered, the spot price is relatively cheaper than the futures price (futures price> spot price). Such a phenomenon is called Contango, and in this case discount occurs when naphtha is purchased.
1 is a graph for explaining naphtha trading using a crack.
Typical cracks are calculated as MOPJ - Brent * 7.5 for Asia. 7.5 above is to match the Brent oil price unit with the MOPJ price unit, in the case of Europe, 7.45 is used. The reason why we see Brent oil besides other crude oil in naphtha trading is that Brent oil market is in Europe and can be observed like NEW and market is opened an hour earlier than WTI (West Texas intermediate) and it is cheaper than WTI . Therefore, the effect of brent oil on naphtha price is relatively larger than that of other crude oil.
Referring to FIG. 1, the naphtha trading method using cracks is a kind of hedging method. When cracks are low, the MOPJ price is relatively lower than the Brent oil (or the Brent oil price is relatively higher than the MOPJ ), So naphtha is purchased at this time, and Brent Yu has a deal contract to sell it. At the same time, when a certain period of time has elapsed (when the crack is expected to peak), the naphtha that was purchased at the bottom is sold, and Brent Yu concludes a purchase reservation for purchase. It will be done. In addition, naphtha cracks are also used to realize profits through linked marketing methods in paper markers when cracks are expected to rise in the future. It is also used as a method of real transactions by fixing cracks.
Principal component analysis refers to the analysis of multivariate data by a number of orthogonal factors that can explain the overall variation of the data without considering all the variables (the first to third principal components, ) Is used for analysis. In other words, it is a technique to reduce the variance of multidimensional variables to the number of enemy number called pricipal component (Pc). Specifically, Principal Component Analysis is an analytical technique that explains the factors represented by the linear combination of these variables using the variance-covariance relationship between several variables, and explains most of the total variance with some of the major components .
2 is a flowchart showing a method of predicting a naphtha crack according to an embodiment of the present invention.
The method of predicting naphtha cracks according to one embodiment of the present invention is a method of predicting cracks which are price differences between Brent oil and Naptha.
The method for predicting naphtha cracks is a method for predicting naphtha cracks by using a principal component analysis technique using eigenvalues and cumulative values of a plurality of independent variables including European gasoline (Mogas) and naphtha price spread (hereinafter also referred to as 'independent variable 1' (A) (S101).
The method for predicting a naphtha crack includes a step (b) (S102) of constructing a regression model for regression analysis of a correlation between a plurality of principal components and a crack to be predicted.
The method for predicting a naphtha crack includes a step (c) (S103) of calculating a principal component and a crack predicted value through a regression model, respectively.
In general, regression analysis is a statistical technique that grasps the correlation between two or more variables. Linear regression analysis is a method of estimating a linear equation between two variables, assuming that there is a linear relationship between the two variables, with one variable as the independent variable and the other as the dependent variable. Simple regression analysis also shows a technique for analyzing the relationship between one independent variable and a dependent variable.
The method for predicting naphtha cracks in accordance with one embodiment of the present invention is configured to predict naphtha cracks through principal component analysis techniques and linear regression analysis.
Also, in step (b), the regression model may be a linear regression model.
FIG. 2 is a graph showing the relationship between the first main component and the crack, FIG. 3 is a graph showing the relationship between the second main component and the crack, and FIG. 4 is a graph showing the relationship between the third main component and the crack.
For example, in step (b), a regression model can be constructed based on the linear relationship between the first main component (Pc1) and the crack and the linear relationship between the second main component (Pc2) and the crack. Here, the crack can be predicted as an average value of the cracks calculated through the linear relationship between the crack generated through the linear relationship of the first main component and the crack and the second main component and the crack .
Thus, 63% of the total crack fluctuation can be explained using the first main component (Pc1) and the second main component (Pc2).
Alternatively, in step (b), the linear relationship between the first main component (Pc1) and the crack, the linear relationship between the second main component (Pc2) and the crack, and the linear relationship between the third main component (Pc3) A regression model can be constructed based on the linear relationship of Here, cracks are predicted by the average value of the cracks calculated through the linear relationship between the first main component and the cracks, the cracks calculated through the linear relationship between the second main component and the cracks, and the third main component and the cracks, .
Thus, 76% of the total crack fluctuation can be explained by using the first main component (Pc1), the second main component (Pc2), and the third main component (Pc3).
First, the multiple independent variables used in principal component analysis are as follows.
Naphtha prices are MOPJ prices. In addition, the plurality of independent variables may include the ratio of the Asian LPG propane price to the Asian LPG butane price (hereinafter also referred to as "
Here, independent variables 1 to 5 are terms used to distinguish independent variables from one another. The independent variables 1 to 5 are data provided externally to the respective states. In addition, these data are all variables affecting Asian naphtha crack variability. For example, if the independent variable 1 increases, naphtha prices in Asia may rise. In addition,
A linear relationship of the first principal component and a plurality of independent variables can be derived and a linear relationship of the second principal component and a plurality of independent variables can be derived through the principal component analysis and the linear relationship of the third principal component and the plurality of independent variables can be derived, Can be derived.
The regression model can be constructed based on the linear relationship between the first main component (Pc1) and the crack, the linear relationship between the second main component (Pc2) and the crack, and the linear relationship between the third main component (Pc3) and the crack.
For example, in the linear relation of the first principal component (Pc1) and a plurality of independent variables, the coefficient of each independent variable is calculated by using a strong linear relationship of monthly spread, ), The scenario for the next month can be constructed. Specifically, it can be calculated by comparing the actual value and the simulation value. From this, it is possible to calculate the predicted crack value by using the linear predictive value of each principal component and the linear regression model between each principal component and the crack. Thereafter, as described above, the cracks calculated through the linear relationship between the first main component and the crack, the cracks calculated through the linear relationship between the second main component and the crack, and the cracks calculated through the linear relationship between the third main component and the crack, Cracks can be predicted with an average value.
For example, if we look at the standardization and volatility of each major component, we can see that naphtha exports and European gasoline cracks decrease as the LPG price increases through the first major component (which accounts for 41% of the total variance) We can see that the premium increases as the supply of naphtha for Asian petrochemicals decreases through the 2 main components (22% of total fluctuations can be explained), and the third component (14% of total fluctuations) And the export of naphtha decreased.
Hereinafter, a naphtha crack prediction apparatus for performing the above-mentioned method of predicting a naphtha crack will be described.
The apparatus for predicting naphtha cracks is a device for predicting a crack, which is a price difference between Brent oil and naphtha.
Specifically, the naphtha crack prediction apparatus includes an input unit into which a plurality of independent variables including a European gasoline (Mogas) and a naphtha price spread (independent variable 1) are input. Here, each independent variable is the same as those described in the method for predicting naphtha cracks, all of which are provided in the corresponding state.
In addition, the naphtha crack prediction apparatus analyzes the input independent variables into a plurality of principal components through principal component analysis using eigenvalues and cumulative values, and calculates a regression model to regress the correlation between a plurality of principal components and predicted cracks And a controller for calculating the principal component and the crack predicted value through a regression model, respectively.
The naphtha crack prediction apparatus may have various terminal forms, such as a computer, on which a program on which the above-described prediction method is performed may be executed.
As described above, the regression model may be a linear regression model.
Further, the control section may be provided to constitute a regression model based on the linear relationship between the first principal component and the crack and the linear relationship between the second principal component and the crack. At this time, the controller may be provided to predict a crack at an average value of cracks calculated through a linear relationship between the crack and the second main component and the crack, which are calculated through the linear relationship of the first main component and the crack.
Alternatively, the control section may be configured to form a regression model based on a linear relationship between the first principal component and the crack, a linear relationship between the second principal component and the crack, and a linear relationship between the third principal component and the crack. At this time, the control section calculates the average value of the cracks calculated through the linear relationship between the first main component and the crack, the cracks calculated through the linear relationship between the second main component and the crack, and the cracks calculated through the linear relationship between the third main component and the crack Can be prepared to predict cracks.
The foregoing description of the preferred embodiments of the present invention has been presented for purposes of illustration and various modifications, additions and substitutions are possible, without departing from the scope and spirit of the invention, And additions should be considered as falling within the scope of the following claims.
Claims (20)
(a) analyzing a plurality of independent variables including a price difference of European gasoline (Mogas) and naphtha into a plurality of principal components through a principal component analysis technique using eigenvalues and cumulative values;
(b) constructing a regression model for regression analysis of correlations between a plurality of principal components and predicted cracks; And
(c) calculating the principal component and the crack predicted value through a regression model, respectively.
In step (b), the regression model is a linear regression model that predicts naphtha cracks.
A method of predicting a naphtha crack in which a regression model is constructed based on a linear relationship between a first principal component and a crack and a linear relationship between a second principal component and a crack in step (b).
A method for predicting cracks in an average value of cracks calculated from a linear relationship between cracks and a crack generated through a linear relationship between a first principal component and a crack and a second principal component and cracks.
A method of predicting a naphtha crack comprising a regression model based on a linear relationship between a first principal component and a crack, a linear relationship between a second principal component and a crack, and a linear relationship between a third principal component and a crack in step (b).
Cracks calculated from the linear relationship between the first main component and the cracks, the cracks calculated from the linear relationship between the second main component and the cracks, and the naphtha cracks having the cracks predicted by the average value of the cracks calculated through the linear relationship between the third principal component and the crack Lt; / RTI >
Naphtha price is the forecast method of naphtha crack, which is MOPJ price.
A plurality of independent variables is a prediction method of naphtha crack including the ratio of Asian LPG propane price and Asian LPG butane price to Asian naphtha price.
Wherein the plurality of independent variables includes at least one of Asian Naphtha premium and European gasoline crack.
Wherein the plurality of independent variables includes a naphtha feed amount for Asian petrifaction.
An input unit for inputting a plurality of independent variables including a price difference between European gasoline (Mogas) and naphtha; And
The input independent variables are analyzed by principal component analysis using eigenvalue and cumulative value, and a regression model is constructed for regression analysis of correlation between multiple principal components and predicted cracks. And a control unit for calculating a predicted value of cracks, respectively.
The regression model is a linear regression model.
And the control section constitutes a regression model based on a linear relationship between the first principal component and the crack and a linear relationship between the second principal component and the crack.
Wherein the control section predicts a crack at an average value of cracks calculated through a linear relationship between a crack and a second main component and a crack calculated through a linear relationship between the first main component and the crack.
The control unit is configured to construct a regression model based on a linear relationship between the first principal component and the crack, a linear relationship between the second principal component and the crack, and a linear relationship between the third principal component and the crack.
The control unit predicts a crack by an average value of the cracks calculated through the linear relationship between the first main component and the crack, the cracks calculated through the linear relationship between the second main component and the crack, and the third main component and the crack, A naphtha crack prediction device.
Naphtha price is MOPJ price, Naphtha crack prediction device.
A plurality of independent variables are naphtha crack prediction devices including the ratio of Asian LPG propane price to Asian LPG butane price compared to Asian naphtha price.
Wherein the plurality of independent variables comprises at least one of an Asian naphtha premium and a European gasoline crack.
Wherein the plurality of independent variables include a feed amount of naphtha for Asian petrifaction.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
KR1020150135481A KR20170036385A (en) | 2015-09-24 | 2015-09-24 | Estimating method and apparatus for crack of Naphtha |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
KR1020150135481A KR20170036385A (en) | 2015-09-24 | 2015-09-24 | Estimating method and apparatus for crack of Naphtha |
Publications (1)
Publication Number | Publication Date |
---|---|
KR20170036385A true KR20170036385A (en) | 2017-04-03 |
Family
ID=58589394
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
KR1020150135481A KR20170036385A (en) | 2015-09-24 | 2015-09-24 | Estimating method and apparatus for crack of Naphtha |
Country Status (1)
Country | Link |
---|---|
KR (1) | KR20170036385A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114414660A (en) * | 2022-03-18 | 2022-04-29 | 盐城工学院 | Method for identifying axle number and cracks of railway vehicle wheel set |
-
2015
- 2015-09-24 KR KR1020150135481A patent/KR20170036385A/en not_active Application Discontinuation
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114414660A (en) * | 2022-03-18 | 2022-04-29 | 盐城工学院 | Method for identifying axle number and cracks of railway vehicle wheel set |
CN114414660B (en) * | 2022-03-18 | 2024-01-12 | 盐城工学院 | Axle number and crack identification method for railway vehicle wheel set |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Gopalan et al. | Duration of executive compensation | |
Ball et al. | Is audit quality impacted by auditor relationships? | |
Héricourt et al. | Exchange rate volatility, financial constraints, and trade: Empirical evidence from Chinese firms | |
Adams et al. | Liquidity constraints and imperfect information in subprime lending | |
Verma et al. | Noise trading and stock market volatility | |
Charles et al. | Market efficiency in the European carbon markets | |
Bonsall IV | The impact of issuer-pay on corporate bond rating properties: Evidence from Moody׳ s and S&P׳ s initial adoptions | |
US8244617B2 (en) | Cash flow rating system | |
Chan et al. | Jumping hedges: An examination of movements in copper spot and futures markets | |
Westgaard et al. | Co-integration of ICE gas oil and crude oil futures | |
Roche et al. | Why does junior put all his eggs in one basket? A potential rational explanation for holding concentrated portfolios | |
Huang et al. | Does corporate derivative use reduce stock price exposure? Evidence from UK firms | |
Etornam | The impact of oil price shocks on the macroeconomy of Ghana | |
Castagnetti et al. | Euro corporate bond risk factors | |
Cifarelli et al. | A dynamic model of hedging and speculation in the commodity futures markets | |
Benos et al. | Funding constraints and liquidity in two-tiered OTC markets | |
Jin et al. | Decision making in trade credit financing: impact of loss aversion and power imbalance | |
KR20180054168A (en) | Estimating Method For Arbitrage of Naphtha | |
KR20170036385A (en) | Estimating method and apparatus for crack of Naphtha | |
Bettendorf et al. | German wage moderation and European imbalances: Feeding the global VAR with theory | |
Kovalev et al. | Dividend policy as a factor for managing company value: Comparing trends in emerging markets | |
Jafarian et al. | Impact of oil price fluctuations on returns of different sectors of Malaysian Stock Market | |
Lezar | Real Exchange Rate of Moroccan Currency: Appreciated or Depreciated? | |
KR20170036386A (en) | Estimating method and apparatus for premium of Naphtha | |
Ahmad et al. | Volatility exchange rate and economic growth: insight from ASEAN Member countries |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
A201 | Request for examination | ||
E902 | Notification of reason for refusal | ||
E601 | Decision to refuse application |