Using an Econometric Approach (VAR model), I will be studying the implications of Scottish independence on the Electricity market and its consumers in Scotland. In other words this dissertation studies whether it’s better for Scotland to function within the UK framework or function independently from an Economic perspective. To answer this, an Econometric Model alone would not be sufficient due to the limitation of data problem, thus results would not be 100% accurate. To overcome this, I will be discussing some of the possible alternatives and scenarios that might occur if the Independence Occurs, and what repercussion each scenario might have.
TABLE OF CONTENTS
LIST OF FIGURES
LIST OF TABLES
CHAPTER ONE
Introduction:
CHAPTER TWO
2. Literature Review:
2.1 Overview of the current UK Electricity Market:
2.2 Previous Studies on the Implications of Scottish Independent on the Electricity Market:
CHAPTER THREE
3. Methodology:
3.1 Choosing an Optimal Lag Length:
3.2 Stationarity of Data Diagnosis:
3.3 Causality Tests and Block Significance:
3.4 Impulse Response:
CHAPTER FOUR
4. Results:
4.1 Unit Root Test Results:
4.2 Granger Casuality in VAR models:
4.3 Impulse Response Function:
CHAPTER FIVE
5. Discussion and Policy Implications:
CHAPTER SIX
6. Conclusion:
APPENDICES
REFERENCES
ABSTRACT
Using an Econometric Approach (VAR model), I will be studying the implications of Scottish independence on the Electricity market and its consumers in Scotland. In other words this dissertation studies whether it’s better for Scotland to function within the UK framework or function independently from an Economic perspective. To answer this, an Econometric Model alone would not be sufficient due to the limitation of data problem, thus results would not be 100% accurate. To overcome this, I will be discussing some of the possible alternatives and scenarios that might occur if the Independence Occurs, and what repercussion each scenario might have.
ACKNOWLEDGEMENT
I would like to thank my Parents, and my sisters for their continuous unlimited support, I will be forever indebted to you.
My Appreciation also goes to Dr. Agathe Rouaix for her guidance, and valuable input throughout the process of writing this dissertation.
LIST OF FIGURES
Figure1. Energy Flow Patterns in the NTS
Figure2. Average Electricity Bill prices in Scotland, England, and Wales
Figure3. Main Stakeholders of the UK Electricity Industry
Figure4. Average Electricity and gas Prices (per kWh) across EU
Figure5. Estimated household energy impacts in Scotland in 2020
Figure6. Power Flow 2015-2030-All Plants
Figure7. An overview of a collaborative Energy Integration scheme
LIST OF TABLES
Table 4.1.1.1 Electricity Consumption in the UK unit root test
Table 4.1.1.2 Electricity Supply of UK unit root test
Table 4.1.1.3 Electricity Prices in UK unit root test
Table 4.1.1.4 Buy Out Price of ROCs unit root test
Table 4.1.1.5 Portion of Electricity Supplied by Renewables in UK unit root test
Table 4.1.2.1 Electricity Consumption of Scotland unit root test
Table 4.1.2.2 Electricity Supply of Scotland unit root test
Table 4.1.2.3 Electricity Prices in Scotland unit root test
Table 4.1.2.4 Buy-out Prices of ROCs in Scotland unit root test
Table 4.1.2.5 Portion of Electricity Supplied by Renewables in Scotland unit root test
Table 4.4.1.1 Overall VAR estimates of model 1
Table 4.4.2.1 Overall VAR estimates of model 2
Table 4.3.1.1 Impulse response functions for the variables of model 1
Table 4.3.2.1 Impulse response functions for the variables of model 2
CHAPTER ONE Introduction:
The collaboration between Scotland and the rest of the United Kingdom has resulted in a robust integration across various areas including the main focus of this dissertation which is Energy and Electricity in specific. In the event of a yes vote in the referendum for independence, Scotland would become a new totally independent state, while the remaining UK members (England, Northern Ireland, and Wales) would continue to exist as before. The disintegration of this Union might have significant changes on the relationship between the UK government and the new independent Scottish State. The Scottish Independence would change all Issues related to Energy in the region, and how the current Electricity generation and transmission framework functions. Thus it’s important to discuss the matter and present a detailed and comprehensive analysis to the voters of the referendum before the voting procedure takes place.
A key aspect in the management of the current electricity network is that the costs of building and maintaining the generation and transmission systems are spread across the whole of UK. The UK Government believes this is a logical method within the current market provisions; drawing on the advantage of having a wider consumer base to help cover the immense costs associated with ensuring the network is suitable for supplying the market with no market shortages. Figure 1 below illustrates how the mechanism of energy flow worked in the UK back in the 90’s and the current flow;
Abbildung in dieser Leseprobe nicht enthalten
Figure1. Energy Flow Patterns in the NTS[1]
Source: National Grid, Gas Ten Year Statement, December 2012.
It is evident that the level of integration and interconnected of Energy flow has been developed tremendously in the twenty years’ time between the 1990’s and 2013. Red Arrows indicating movement of Energy were indicating a unidirectional movement from some regions into others, but developed grids and the sophistication of the current energy network in the UK have changed this relationship to become bi-directional (Dahl, 2008).
Thus when we discuss the prospect of Independence in Scotland a measure its effects on the Electricity Market a lot of issues must be taken into account. Consumer choice and behavior will be affected due to changes in Market Dynamics. Electricity price volatility is also a threat to be mentioned when discussing this because the security of supply will be in jeopardy at this moment. Will Scotland be able to supply its Electricity enough for internal use and for exporting purposes, in light of the current renewable obligation schemes (ROCs) and incentives for companies?
Thus, this dissertation examines the relationship between consumers (households) consumption, and various variables related to the Electricity market in the United Kingdom. I have chosen the following independent variables; Electricity Supplied, Portion of electricity generated by renewables, buy out price of ROCs, and Electricity prices. Two models were constructed to study this relationship, the first model takes into account the data related to these variables with respect to the entire United Kingdom, and the second one only studies the Scottish electricity market. The purpose of constructing these models is to observe how these various interact with each other, and to see to what extent a change to the current structure of the electricity market might affect the Scottish consumer. In this dissertation I will not apply any forecasting techniques since it’s difficult to anticipate what the referendum might yield, and forecasting qualitative data will provide us with meaningless results. Instead I will be studying the impulse response functions of the variables and their reaction graphs.
This dissertation is divided into six chapters, chapter 2 is the Literature review that is divided into two sections; the first provides the reader with an overview of the current Electricity market in the UK. The second part discusses some of the previous studies made on the topic of independence, and provides some insight on some projections of what might happen to the Electricity market in Scotland is declared as an independent state. Chapter 3 explains in depth the methodology used to obtain the results in Chapter 4, it discusses the theory behind the choice of the optimal lag length, carrying out unit root tests and granger causality tests, and the importance of Impulse response functions. Chapter 4 discusses and explains the results we have obtained from the tests we have obtained, and how we treated the data. In chapter 5 we introduce some of the implications and possible scenarios that might happen if Scotland is declared as an independent entity based on the results we have obtained and previous literature, followed by conclusion, where we sum our findings in Chapter 6.
CHAPTER TWO 2. Literature Review:
This chapter is divided into two sections; the first one provides a general overview of the current UK electricity market, its history, milestones, and its method of operation. The second part of the Review discusses some previous studies conducted on the issue of independence, discussion of some hypothetical scenarios that might happen, providing a useful insight on what could happen.
2.1 Overview of the current UK Electricity Market:
The Current UK electricity market is a fully Liberalized and privatized market; the process of Liberalization of Electricity markets started in the mid-1980. The Energy Act of 1983, made the supply market open beyond the existing 12 areas present at the time (Bourn, 2006). The Energy Act facilitated the creation of a wholesale market where electricity generators can sell electricity in spot markets to match demand and supply (99/02050 Electricity market liberalization and environmental performance: Norway and the UK, 1999).
Even though Liberalization took place in 1980, a single integrated market in the UK was introduced 20 years later in 2005, when (BETTA) or the British Electricity Trading Transmission Arrangements was established. Before 2005 Scotland had its own separate electricity market which was dominated by two electricity firms, Scottish Power, and Southern Energy. The two companies controlled generation, transmission, and supply. They controlled around 74 percent of supply to domestic consumers in Scotland, and around 76 percent of the generation capacity (Gov.uk, 2017). The presence of only two firms in the market, barriers to entry where many and small producers could not compete in the market due to high costs of production (Sioshansi and Pfaffenberger, 2006). All of this led to the monopolization of the Scottish Electricity market, which ultimately led to limiting wholesale competition, eliminating choices offered to the consumer, and increased prices. The high levels of market concentration in the Scottish market for electricity, was leading to obvious market inefficiencies, as consumer electricity bills were 30 pounds higher on average than those nationwide (England and Wales). Before the single integrated market was introduced, electricity was traded between Scotland and the rest if the UK through inter-connectors (The competitive electricity market from 1998: Scottish trading arrangements September 1997, 1997). Which meant that Scottish Electricity had a limited capacity to load on the national grid, and the Scottish were obliged to pay trading fees for this process to be done, all fees were abolished after the BETTA agreement (The impact of BETTA on the Settlement Agreement for Scotland (SAS), 2003).
The movement of Scotland to this bigger more diverse market after 2005, provided consumers with greater choice, lower prices, and allowed them to enjoy the perks of increased competition among suppliers (Scottish Law Commission (Scot. Law Com. No. 49). Electricity (Scotland) Bill. Report on the consolidation of certain enactments relating to electricity in Scotland. (2007). Cambridge [England]: Proquest LLC, pp.78-96.). Companies were incentivized to provide sustainable and dependable supplies to attract and retain customers, stabilizing prices even more. Nine new suppliers have entered the Electricity market since 2010 in various regions around the UK; some new small suppliers were among those new suppliers who are having an increased influence in this new Electricity Market (Jalali and Kazemi, 2015). Figure 2 below illustrates the drop in Consumer Electricity bills after the BETTA agreement was introduced in England, Wales, and Scotland, which reflects the positive impacts the BETTA agreement, had on average Electricity bill prices, which has dropped significantly in the post BETTA era as shown.
Abbildung in dieser Leseprobe nicht enthalten
Figure2. Average Electricity Bill prices in Scotland, England, and Wales
Source: Quarterly Energy Prices: March 2014, Department of Energy and Climate Change, March 2014, Table 2.2.2.
We have to take a close look on the current framework of operation of the electricity market in the UK, to better understand the size of the implications the potential independence might have on the current Scottish electricity in specific and the UK marker as a whole. Presently the single market is operated by one regulator which is the Office of Gas and Electricity Markets (Ofgem), and one system operator which is the National Grid. As a system operator the National Grid manages the electricity supply and demand throughout Great Britain, on the other hand as a regulator Ofgem role is to control and monitor prices, and is responsible for protecting consumers against market inefficiencies, by ensuring that wholesale and supply markets are competitive. Ofgem has introduced its Retail Market Reform, which is designed to address consumer needs and rights, and ensure that energy companies place consumers on the cheapest tariff with respect to their preferences. Figure 3 below provides an overview for the GB electricity industry mechanism and main stakeholders;
Abbildung in dieser Leseprobe nicht enthalten
Figure3. Main Stakeholders of the UK Electricity Industry
This method of operation allows the UK citizens to benefit from advantages of collective decision-making, as it reduces price discrepancies, and increases risk sharing. For example, Scotland is set to benefit from a six billion of investment in the next 15 years. This investment is heavily and mostly linked to gas and electricity networks, and will further strengthen Scotland’s access to the entire GB network (RIIO-T1: Final Proposals for SP Transmission Ltd and Scottish Hydro Electric Transmission Ltd, 2012). Also in 2013 around 132.5 million pounds were invested in Scotland’s network upgrades, this amount when spread on all UK citizens poses a much lower burden on Scottish people, rather than being responsible alone for the upgrades (Park, 2013). Questions remain on whether such investment inflows will continue to finance Scottish Energy projects if independence transpires? And if the Scottish government will be able to devise a new efficient mechanism of electricity generation and sale as the one implemented at the moment?
These questions seem sensible to ask at the moment, because Integration of the Electricity market in the UK, not only affected consumers and suppliers but also enhanced the overall energy Network in the Region. This developed integrated network allowed for an effective management of a varied energy mix including renewables[2], nuclear, and coal to meet the projected demand nationwide (Boyle, 2012). The result of this Integration is evident when comparing the average prices of Electricity and Gas across Europe. A key feature of this system is that the costs incurred by the British government to construct the infrastructure of the National Grid and various transaction costs are spread across the Whole of Great Britain. This is exceptionally important for Scotland, because for example all GB consumers are sharing the costs of the network upgrades taking place in Scotland, and not only Scottish residents. This is shielding Scottish consumers living in remote communities from incurring the high costs associated with electricity supply. Figure 4 illustrates how this effective management allowed the UK ranked among the cheapest European countries in consumer Electricity and gas prices.
Abbildung in dieser Leseprobe nicht enthalten
Figure4. Average Electricity and gas Prices (per kWh) across EU.
Source: Eurostat data, published in DECC’s Quarterly Energy Prices for the period of January to June 2013. Gas data for Finland was unavailable for this period. Data sorted by electricity prices
Ofgem has introduced the Renewables Obligation Certificates system, known as the ROCs in 2002. This resembles a trading scheme by which electricity suppliers in England, Wales, and Scotland are allowed to purchase renewable electricity in order to meet targets set out in the government Renewables Obligation. Each ROC archives the details of how the unit of electricity was produced, who produced it and who bought it. ROCs form part of a trading scheme that enables firms to receive payment for the electricity they produce. The scheme is open to anyone who owns a system that is 50kW in size or larger. In other words a ROC[3] is a certificate that is issued for every MW/h of renewable electricity you generate from your on-site system. The ROCs system is important when discussing the Electricity market because the majority of Scotland’s Electricity is produced using Wind farms, and that will be under influence if Independence takes place (The EU's target for renewable energy, 2008). It’s worth the mention that most probably ongoing ROCs will be honored by the British government if independence occurred, however future ROCs must be covered by the Scottish government. Given the size of the Scottish economy and the dependence of the Scottish electricity market on the overall British market, a possible disintegration in the market may result in a sharp Increase in the price of energy sources in Scotland (Retd, 2013). All post-independence buy out prices of ROCs will increase in the Scottish territories due to the loss of cooperation with the regional market regulators; the national grid and Ofgem (Peeters and Schomerus, 2014).
2.2 Previous Studies on the Implications of Scottish Independent on the Electricity Market:
Few studies have been made on the Implications of Scottish independence on the Current Electricity market in the UK. Nonetheless it is important to examine these studies to evaluate the possible scenarios that might happen if the Independence takes place, and what effects (positive or negative) they will carry.
Grant, McGregor, and Swales have raised this question when they circulated their study “Independence and the market for electricity in Scotland” in 2017. They were skeptical about the possible outcomes of the independence (Swales, McGregor and Allan, 2017). Since the Electricity market in Scotland and the UK as a whole is a liberalized market as we have mentioned earlier, the impact of any constitutional change would depend on the way different stakeholders (i.e. produces and consumers) are going to react in terms of supply and demand. In addition we have to take into account that Electricity cannot be stored easily, and here a question of the method of governance Scottish policymakers are going to adopt must be asked. The authors have deduced in conclusion that the main focus of the Scottish government should surely remain the security of supply, when they make changes to the Electricity related matters in the constitution (Swales, McGregor and Allan, 2008). Electricity inflows from England and wales will surely decline after the independence, where new taxation and tariff schemes will be introduced.
Security of supply would typically require a balanced portfolio of generation technologies of sufficient capacity; the reliability of these technologies; and resilience to shocks (including shocks to supplies as well as to prices) (Security of electricity supply, 2000). Unfortunately my model doesn’t take security of supply as one of the variables, even-though it is an important factor to study the implications of independence on the availability of electricity to Scottish consumers. The reason behind the exclusion of this variable is the fact that it’s hard to quantify or measuring in numbers thus we cannot in cooperate it in our model efficiently and obtain meaningful results.
The current electricity grid covering Great Britain after the BETTA agreement, linked through interconnectors with Europe and Ireland, relates the spatial dimension to security of electricity supply directly to the size of the grid (Swales, McGregor and Allan, 2017). Within a single physically integrated electricity market spanning over Great Britain, for example, it would seem appropriate to consider security of supply at the whole-grid level. However, security of supply becomes a rather more difficult objective to achieve, other things being equal; the smaller is an economy; the more concentrated is its portfolio of electricity generators; the more limited is the market (in terms of capacity to trade electricity, and consumer choice); the smaller is storage capacity; the more restrictive is its transmission and distribution system; and the more geographically isolated are its residents (Hogan, 2012).
A different analysis conducted by the UK Government that studied the proposed scenario of Scottish independence. The study shows that even in a hypothetical scenario of independence of Scotland where there were no flows of electricity between an independent Scottish state and the continuing UK there would be only a small impact on capacity margins in England and Wales. This is because the lack of access to Scottish-generated electricity under such a scenario would be largely balanced by the removal of Scottish peak demand from the system, so would not significantly heighten the risk of blackout in GB, in other words the national grid would be able to adjust to the change in demand patterns accordingly. The intrinsic features of the GB market mean that more costly forms of renewables generation in Scotland can be met automatically through unimpeded access to the GB consumer market (Anon, 2017). This means that Scotland is the entity that would be immensely affected by independence, while other states in the union will continue to function normally under current rules and regulations. For instance one of the draw backs of independence is that many remote areas in Scotland can no longer benefit from pooled investment in the transmission network, as Independent means also the gradual withdrawal of the BETTA agreement(Hogan, 2012)..
Nonetheless, in order to provide estimates of potential capacity margins under the scenario of separate markets and in the hypothetical absence of any electricity flows between Scotland and the rest of the UK, modeling has been undertaken based on the projected 2020 generation mix in GB from the Electricity Market Reform Delivery Plan analysis constructed Ofgem. The outputs are dependent on the following assumptions (Anon, 2017). The modeling approach assumes that Scotland achieves its target of 100 per cent renewable electricity as a share of gross annual consumption in 2020, that de-rated capacity in 2020 is 6.4GW, with peak demand of 5.5GW. In the rest of GB, de-rated capacity and peak demand are estimated to be 56GW and 51GW in 2020 respectively. The projected changes in prices for Scottish consumer are illustrated in Figure 5 with low-high price margins.
Abbildung in dieser Leseprobe nicht enthalten
Figure5. Estimated household energy impacts in Scotland in 2020.
Source: DECC Modeling.
The Electricity Market Reform Delivery Plan estimated the de-rated capacity margin[4] for GB to be 8.3 percent in 2020. This estimation is based on the assumption that current integrated market arrangements remain and that the grid continues to be managed to ensure the supply and demand of electricity is balanced for UK as a whole, and according to the same regulations. Dividing capacity margins for Scotland and the rest of UK based in the assumption of capacity and peak demand (Anon, 2017), results in a 7.4 per cent de-rated capacity margin for the rest of UK and a 16.3 per cent de-rated capacity margin for Scotland in 2020. This demonstrates that if there were no flows of electricity between Scotland and England or Northern Ireland in 2020, it would not have a significant impact on the capacity margin in the rest of UK. This is because the assumed loss of Scottish de-rated capacity to the rest of UK in 2020 is largely offset by the removal of the need to meet Scottish peak demand (Gov.scot, 2017). The Scottish government offered a 2015-2030 projection scenario, where all electricity production using renewables is produced in Scotland; the results are shown in Figure 6;
Abbildung in dieser Leseprobe nicht enthalten
Figure6. Power Flow 2015-2030-All Plants.
Source: Gov.scot, 2017.
The generation scheme proposed in the figure illustrates that excess Scottish generation will occur before 2020, because Scotland is benefiting from the UK network grid for electricity generation.
From this Scenario it’s important to deduce the following points:
- Scotland will be exporting most of its generated electricity, which imposes security of supply issues, if this persists after independence.
- By 2020 additional Scottish generation will lead to additional export capacity required, which means at least one additional link will be required by 2020 (the shaded area on the chart)
All in all and after discussing the dynamics of the current UK electricity market and previous studies made on Scottish independent we can see that the matter of independence should be dealt with great care. In chapter 5, we discuss some possible policies and recommendation that should be taken into account to make this political endeavor as successful as possible, and to overcome some of the technologies presented here and in the results chapter.
CHAPTER THREE 3. Methodology:
This dissertation investigates the effects of Scottish independence on Scottish Electricity Market and Consumers. This will be accomplished through building two vector auto-regressive (VAR) models to study the relationship between the following factors; Electricity consumption of consumers and households (dependent variable), Electricity Supplied (independent variable), Portion of electricity generated by renewables (independent variable), buy out price of ROCs (independent variable), and Electricity prices (independent variable). The choice of building two models was based on the intuition that we have to study the UK electricity market as a single integrated market. Then build a second model to study the Scottish model by itself, to monitor the interaction of the variables to change. The frequency of the data used for the model is yearly data spanning over the period 1970-2017, producing 48 raw observations. The data is collected from several resources[5] since it’s hard to obtain all the data needed from the same source, where each data source focused on a specific variable In their study.
The choice of the VAR model is based on two solid reasons; the first one is that the limited amount of observations we have doesn’t allow the use of a Vector Error Correction model (VECM) to model the data. A VECM model is more suitable for studying the Long Run relationships among the variables at hand, however we will discuss the issue of the studying the LR relationship later In this chapter. The second reason behind our choice of the VAR model is its ability to capture the complex dynamics of the data in multiple time series. The following system of equation, will demonstrate how a VAR model is able to express the dynamics of the model where each variable has its own equation in terms of the other variables (representing the inter-relational characteristics of the data);
illustration not visible in this excerpt
Where 1 is the Energy consumption in UK, 2 is the Energy Supply by the UK, 3 is the Electricity price, 4 Buy out price of ROCs, 5 is Portion of Electricity Supplied by Renewables in UK, and i is the white noise term, in Model 1.
Where 1 is the Energy consumption in Scotland, 2 is the Energy Supply by Scotland, 3 is the Electricity price in Scotland, 4 Buy out price of ROCs, 5 Portion of Electricity Supplied by Renewables in Scotland, and i is the white noise term in Model 2.
3.1 Choosing an Optimal Lag Length:
It’s crucial to discuss next the optimal lag length to be used in our model, as lag length represents the optimal number of lags for each variable to be embodied in each model. To deduce this using the E-views software, we use the Information Criteria; the Information Criteria contains two parameters: a term which represents the residual sum of squares (RSS)[6], and a term that acts as penalty for the loss of degrees of freedom as a result of adding extra variables. Adding one more lag, will increase the penalty value, but decrease the RSS. The lag length with the lowest Information criteria Value will be the optimal lag length.
There are several Information Criteria’s that differ only in their assessment of the rigidity of the penalty term. In this dissertation we will be using the two most common information Criteria’s the Schwarz Information Criteria (SIC), and the Akaike Information Criteria (AIC) represented by the Equations below;
illustration not visible in this excerpt
Where log |∑u(p)|, Is the Variance-Covariance Matrix to calculate the RSS, and the penalty Term is calculated by including the number of parameter K, and the number of parameters T as shown above (Kilian,2001).
Because of the know Characteristics of the Schwarz and Akaike information criteria, it’s usually expected that AIC would suggest model with a large number of lags, while the SIC would suggest different order of lags of the model, for different samples taken from the population. However and despite these differences we cannot propose that one of the criteria’s is better than the other. Since the SIC has a more strict penalty term than the AIC, making it more consistent but less efficient at the same time due to the inclusion of larger number of lags.
3.2 Stationarity of Data Diagnosis:
Before estimating and interpreting the results of the models, the stationarity of the data should be tested. Modeling this non-stationary data will yield a spurious regression. A spurious Regression provides misleading statistical evidence of a linear Relationship between independent non-stationary variables. Non-stationary data provides and produce meaningless estimates, which violates the standard OLS regression assumptions.
To detect and overcome the issue of non-stationarity in the data, we conduct the Augmented Dickey Fuller (ADF) Test, and the Philips-Peron (PP) Test for each variable we have individually. The two tests suggest the same null hypothesis which states that the series contains and unit root (non-stationary), the alternative hypothesis suggests that the stationarity of the data. Since ADF and PP tests are one sided tests, and they have a null hypothesis that suggests non-stationarity we can say that they do not follow a standard t-distribution.
Given the Nature of the two tests, the null hypothesis can be rejected (implies stationarity), when the test statistic is more negative than the suggested DF critical values by the tests.
The following model is the model that is going to be tested for Stationarity;
illustration not visible in this excerpt
Where P is the number of lags determined by the information criteria; either the AIC or SIC, ut is the white noise term, 𝛼0 in added to treat the case of flat data that slowly turn around non-zero values, and 𝛽𝑡 which accounts for the time trend if the data is observed to move around a trend line.
Treating non-stationary data and transforming it into stationary data can be done by taking the first difference of the series. This is mathematically done by subtracting yt-1 from both sides of the above equation yielding the following equation:
illustration not visible in this excerpt
Δyt is our new Stationary variable. It’s important to note here that some non-stationary variables may contain more than one unite root, let’s say d roots. The series needs to be differenced d times to become stationary.
Unfortunately using the differencing method to treat non-stationary data might affect any Long Run relationship we have between our variables. Thus we have to test for co-integration before taking the first difference. Co-integration exists when two or more non-stationary variables can be linearly combined and this combination might results in stationary process. Testing for Co-integration among non-stationary series can be conducted through the Johansen test for Co-integration, the choice of this test is based on its flexibility in identifying more than one co-integration relationship. The johansen test is computed according to the following systems; let’s assume we have p variables that we suspect are co-integrated. We ensure first all are of the same order of non-stationarity, most probably of order 1 {i.e. I (1)}, as it’s very unlikely for variables to be of a higher order of integration. Stack the variables that we need to test in a p-dimensional vector and call it yt, and then construct a vector of Px1 of first differences Δ𝑦𝑡 and estimate the VAR according to the following equation:
illustration not visible in this excerpt
Then we test the rank of the matrix π, if it’s equal to Zero this means all eigenvalues are not significantly different from zero, no cointegration exists, otherwise the value of the rank π will correspond to the number of the cointegration vectors.
[...]
[1] National Transmission System
[2] 33.3% of Scotland’s Electricity Is produced using Wind Energy (Scots open Europe's largest wind farm with plans to expand, 2009).
[3] We are interested with the Buy-out Price of the ROC, not the initial price presented in auctions as this is considered a sunk cost.
[4] De-rated capacity margin represents the availability of plant at peak time, specific to each generating technology.
[5] The resources are OFEGM (Office of Gas and Electricity Markets), and the British department for business and industrial strategy.
[6] RSS is the sum of the squares of residuals (deviations predicted from actual empirical values of data). It is a measure of the discrepancy between the data and an estimation model. A small RSS indicates a tight fit of the model to the data (Draper and Smith, 1998).
-
-
-
-
Upload your own papers! Earn money and win an iPhone X. -
-
Upload your own papers! Earn money and win an iPhone X. -
-
Upload your own papers! Earn money and win an iPhone X. -
-
Upload your own papers! Earn money and win an iPhone X. -
-
Upload your own papers! Earn money and win an iPhone X. -
-
Upload your own papers! Earn money and win an iPhone X. -
-
Upload your own papers! Earn money and win an iPhone X. -
-
Upload your own papers! Earn money and win an iPhone X. -
-
Upload your own papers! Earn money and win an iPhone X. -
-
Upload your own papers! Earn money and win an iPhone X. -
-
Upload your own papers! Earn money and win an iPhone X. -
-
Upload your own papers! Earn money and win an iPhone X. -
-
Upload your own papers! Earn money and win an iPhone X. -
-
Upload your own papers! Earn money and win an iPhone X. -
-
Upload your own papers! Earn money and win an iPhone X.