American Interdisciplinary Journal of Business and Economics ISSN: 2837-1909| Impact Factor : 6.71 Volume. 11, Number 3; July-September, 2024; Published By: Scientific and Academic Development Institute (SADI) 8933 Willis Ave Los Angeles, California https://sadijournals.org/index.php/AIJBE| editorial@sadijournals.org American Interdisciplinary Journal of Business and Economics | 35 https://sadijournals.org/index.php/AIJBE EMPIRICAL EVALUATION OF OIL PRICE VOLATILITY AND STOCK MARKET RETURNS IN NIGERIA Ikemenogo, Eze Solomon Department of Economics, Faculty of Social Sciences, University of Lagos, Akoka, Nigeria DOI: https://doi.org/10.5281/zenodo.13236955 Abstract: The study is on empirical evaluation of the impact of oil prices volatility on stock market returns in Nigeria from 1986 to 2021. With the understanding that energy runs like a bloodstream of any business of which oil is a major source, it becomes imperative to explore the incessant oil price fluctuations and its effects on critical indicators of the economy like stock market performance. The objectives include to examine the patterns of oil price volatility and stock market returns in Nigeria, analyse the influence of oil price fluctuations on Nigeria's economic growth, and investigate the causal relationship between oil price volatility and the stock market returns in Nigeria. Arbitrage Pricing Theory (APT) was employed as the framework while the methodology adopted the all-share indices (ASI) as the dependent variable, then, oil prices (OPRICE), exchange rate (EXCH), inflation rate (INFL), gross domestic growth rate (GDPGR), and interest rate (INT) as explanatory variables. Secondary data sourced from UNCTAD, etc. were analysed with aid of EVIEWS 2021. The Generalised Autoregressive Conditional Heteroscedasticity (GARCH) and Granger causality were the estimation techniques used. The study used unit root test to check the stationarity of the variables, the ARDL bound test to check for long-term relationships between the variables, and the ARDL model to estimate both short- and long-term relationships between the variables. A normality test revealed that the study's variables are all typical. Breuseeh-Gdfrey serial correlation revealed no association between the study's variables. The heteroscedasticity test showed there was no outlier’s effect on the output of the result. It was found that inflation and exchange rate volatility are positive and statistically significant; inflation and interest rate are equally positive related to private consumption during the period under investigation. It found out inflation, oil price, exchange rate, and real gross domestic product have positive effect on stock market performance in Nigeria. The work concludes oil price is a major determinant of economic growth in Nigeria. It was recommended that all brokerage firms and investment advisors need to conduct periodic research on macroeconomic environment and advise their clients accordingly on the best counters to invest in owing to the various influences by macroeconomic environment on the stock market performance. Keywords: Oil, Prices, Volatility, Stock, Market, Returns Ikemenogo, Eze Solomon (2024) American Interdisciplinary Journal of Business and Economics | 36 https://sadijournals.org/index.php/AIJBE INTRODUCTION 1.1 Background of the study Oil plays a significant role in the global economy, hence there has been a lot of research done to understand the financial and economic effects of fluctuations in oil prices. There is controversy about the relationship between the oil price and the stock market returns, making it an unpredictable factor in economic growth (Ndlovu, 2019). Oil is a major source of energy globally and a crucial element of Nigeria's economy, significantly influencing its economic and political landscape. Although Nigeria’s oil industry dates to the early 20th century, it was not until after the Nigerian Civil War (1967-1970) that oil began to dominate the economic sphere. The discovery of crude oil has had both positive and negative impacts on Nigeria's economy. The oil sector contributes approximately 90% of Nigeria’s total revenue. It generates employment opportunities, boosts foreign exchange reserves, and supplies energy to various industries and commerce. Nigeria joined the Organization of Petroleum Exporting Countries (OPEC) in 1971 and established the Nigerian National Petroleum Corporation (NNPC) in 1977, a state- controlled entity involved in both the upstream and downstream sectors (Blair, 1976), following the discovery of crude oil by Shell D’Arcy Petroleum, pioneer production commenced in 1958 from an oil field in Oloibiri, Eastern Niger Delta. On the negative side, oil exploitation has led to significant environmental degradation in surrounding communities. This has caused a loss of livelihood and other economic and social challenges. Despite NNPC’s efforts to maximize capacity and improve petroleum product distribution, inefficiencies persist, leading to inconsistent supply and allocation issues. Crude oil prices have experienced significant volatility over time. According to Olayungbo and Ojeyinka (2021), the first major global oil price shock occurred from 1973 to 1974, when prices surged from $3 to $12 per barrel due to the Arabian embargo. The second shock occurred between 1978 and 1979 during the Iranian Revolution, with prices rising from $12 to $18 per barrel. The third shock took place during the Iraq-Iran conflict (19801990), which saw oil prices increase from $28 to $40 per barrel. The 2008-2009 global financial crisis caused oil prices to plummet from $100 to $47 per barrel. The Nigerian Stock Exchange (NSE), established in 1960 and renamed in 1977, is a critical institution in Nigeria's capital market. It has branches in major cities, with its head office in Lagos and another office in Abuja (Akigbo, 1996). The NSE opened in 1961 with 19 listed equities. Today, it lists 328 securities with a total market value of approximately N28.26 trillion as of January 2020 (Akigbo, 1996). The NSE and the Securities & Exchange Commission (SEC), which enforces the Investments & Securities Decree 1999, regulate transactions on the Exchange. The deregulation of Nigeria's capital market in 1993 and the repeal of foreign participation restrictions in 1995 allowed foreigners to engage as operators and investors. Since 1987, the NSE has been a part of the Reuters Electronic Contributor System, facilitating global distribution of stock market data, trade statistics, and company news. The primary objectives of the NSE include developing a system for capital formation, offering efficient resource allocation, providing unique financing options, maintaining market discipline, and broadening share ownership (Akigbo, 1996). The influence of oil price volatility on the capital market is significant. Oil price fluctuations impact the stock market by reflecting the market's expectations of future profitability (Akigbo, 2014). The stock market absorbs the current and expected future impacts of oil price shocks, which are reflected in stock prices and returns. With oil production and exportation playing a crucial role in driving economic growth and development, the volatility of oil prices has been a subject of great importance in the country's Ikemenogo, Eze Solomon (2024) American Interdisciplinary Journal of Business and Economics | 37 https://sadijournals.org/index.php/AIJBE financial landscape (Wang et al, 2022). Over the period from 1990 to 2022, Nigeria experienced notable fluctuations in oil prices, which, in turn, had a profound impact on the performance of its stock market. The effect of oil price volatility on the economy is complex and unsettled, with no consensus on the relationship between financial variables and oil prices. Researchers like Salisu and Oloko (2015), Babatunde et al. (2013), and Fowowe (2013) argue that there is a direct relationship between oil prices and stock market performance. This study aims to link these variables, examining the impact of oil price volatility on stock market returns using disaggregated data. The Nigerian Stock Exchange serves as a catalyst for mobilizing and utilizing private and public savings for productive uses (Akigbo, 1996). Fluctuations in oil prices can pose threats to the stock market, given the oil industry's critical role in providing foreign exchange and total revenue for Nigeria's socio-political and economic wellbeing. 1.2 Statement of the Problem Oil discovery in Nigeria has driven economic growth and infrastructural development but has also caused significant environmental harm, particularly water pollution that has devastated aquatic life and stripped local fishermen of their livelihood. As a mono-economy heavily reliant on oil, Nigeria's macroeconomic stability is highly sensitive to fluctuations in oil demand and supply. The 1970s oil boom led to the neglect of non-oil revenues, an expanding public sector, and poor financial discipline, exposing Nigeria to oil price volatility and financial instability. Despite the efforts by various governments to revive agriculture and diversify the economy which include initiatives like Operation Feed the Nation (OFN), Green Revolution, and National Economic Empowerment and Development Strategy (NEEDS), etc., the economy remains vulnerable. Nigeria's open economy is highly susceptible to crises, which significantly impact its volatile stock market, posing challenges for investors and financial analysts. The Nigerian stock market experiences high volatility due to risks and price shocks. Risk-averse investors often avoid the market due to the uncertainty and volatility in expected returns (Ashamu et al., 2017). High volatility increases unfavourable market premiums, and investors demand higher returns on investments (Atoi, 2014). Oil price fluctuations induce large variations in stock market returns, making oil prices a significant concern for scholars. Most research focuses on developed economies, with limited studies on developing economies like Nigeria, despite its significant stock exchange. 1.3 Objectives of the Study The broad objective of this research study is to examine the causality between oil price volatility and stock market returns. To achieve the broad objectives, it is pertinent to streamline the specific objectives which are to: i) Determine effect of oil price fluctuation on the economic growth in Nigeria. ii) Establish the nexus trend between oil price volatility and stock market in Nigeria. iii) Determine the impact of oil price volatility on stock market returns in Nigeria. 1.4 Research Questions In line with the research objectives, the study aims to provide the answers to the following research questions. i. i. Does oil price volatility have an impact on stock market returns? ii. What is the effect of oil price fluctuation on the economic growth in Nigeria? iii. Is there any nexus between oil price fluctuations and economic growth in Nigeria? 1.5 Research Hypothesis To carry out this study, the following hypothesis must be tested: H01. Oil price volatility has no significant impact on stock market returns. Ikemenogo, Eze Solomon (2024) American Interdisciplinary Journal of Business and Economics | 38 https://sadijournals.org/index.php/AIJBE H02. There is no significant effect of oil price volatility on stock market returns. H03. There is no significant nexus of oil price fluctuation and economic growth. 1.6 Scope of the Study This research study focused on oil price volatility effect and stock market returns in Nigeria from 1986- 2021. The study starts in 1986, a year prior to 1987 because there was a world stock market crash called ‘black Monday’ with worldwide losses of about US$1.71trillion and this significantly affected the world’s market, and it ends with 2021 because of the difficulty in finding accurate and verified data. The study will not cover other African countries who may be facing similar challenges. But the findings of this study can be used as a guide. LITERATURE REVIEW 2.1 Conceptual Review Oil price changes affect numerous economic variables such as interest rates, investment decisions, economic growth, investors' confidence etc. these variables have been documented to affect both the stock market and exchange rate market (Hamilton, 1983; Amano & Van Norden, 1995). Again, oil prices are expressed in US dollars in the international market; hence, the dollar exchange rate may affect the price perceived by oil producing nations (Roubaud & Arouri, 2018). This study reviews the connection between these variables (Oil price, exchange rate and stock market returns) using a bivariate and multivariate approach. To some economists, there is a positive correlation between crude oil price and stock market performance (Cong, Weiy, Jiao & Fan, 2008; Boyer & Filion, 2007; Sadorsky, 2001). According to Babatunde (2013) while the initial effects were contained due to low levels of exposure to complex financial instruments, the large swings in oil prices, combined with the resulting depreciation of the naira and drop in investor confidence led to growing pressures about the oil price volatility which was supported with exposure of oil subsidy frauds by importers of fuel products, who had a high foreign currency obligations owing to the high fuel prices in 2008, the subsequent falling oil prices and devaluations of naira added to the shocks experienced in the Nigerian stock market. 2.1.1 Oil Price and Exchange Rate Theories generally posit that crude oil prices and exchange rates are positively correlated in oil-exporting countries, with higher crude oil prices leading to currency appreciation and vice versa. Crude oil price shocks affect exchange rates through two primary channels: the terms of trade channel and the wealth effect channel. The terms of trade channel suggests that a negative shock to the terms of trade drives down the price of non- tradable goods in the domestic economy, causing the real exchange rate in oil-exporting economies to depreciate and vice versa. The wealth effect channel indicates that a drop in crude oil prices results in losses for oil exporters but gains for oil importers, shifting current account balances and leading to portfolio reallocations between oil trading companies (Akigbo, 2014). A negative oil price shock transfers wealth from oil exporters to oil importers, or higher oil prices lead to higher production costs and inflation, which have contractionary effects on the economy and trade balances. To restore or improve trade balances, the exchange rate must adjust. The impact of oil prices on exchange rates can vary between advanced and emerging market economies. Hamilton (2009) notes that oil price shocks respond directly to economic or geopolitical events and economic downturns (demand-side shocks). Supply-side shocks are driven by disruptions in oil production, such as the Iranian invasion of the U.S. embassy in 1978, the Iraq invasion of Ikemenogo, Eze Solomon (2024) American Interdisciplinary Journal of Business and Economics | 39 https://sadijournals.org/index.php/AIJBE Kuwait in 1980, the Arab Spring in 2000, and the Iranian attack on Saudi oil fields in 2019. These events disrupted oil production without corresponding reductions in demand, driving up prices. Demand-side shocks, on the other hand, are influenced by global economic movements. For example, the economic growth in China and other developing economies significantly increased oil demand without a matching supply increase, leading to high oil prices. Conversely, during the 2007-2009 global financial crisis, a dramatic reduction in oil demand led to a collapse in oil prices. Kilian (2009) identifies three types of oil price shocks: supply-side shocks, aggregate demand shocks, and precautionary demand shocks. Geopolitical unrest often triggers precautionary demand shocks, causing uncertainty about future oil availability and driving up prices. For instance, geopolitical events lead economic agents to expect shortages in oil supply, which results in high oil prices. According to Salisu and Oloko (2015), crude oil price shocks affect exchange rates through the terms of trade and wealth effect channels. A negative oil price shock transforms wealth from oil exporters to importers, leading to higher production costs and inflation, which adversely affects trade balances. To improve trade balance, the exchange rate must adjust. The relationship between exchange rates and oil prices can differ between advanced and emerging market economies. 2.1.2 Exchange Rate and Stock Market Returns Aruori (2011) emphasized that the relationship between the movement of exchange rate and stock returns could be explained with several perspectives. The flow-oriented model of exchange rate behaviour posits that depreciation in the exchange rate for instance would lead to improved trade balance as exports become cheaper. This would result in upward shift in aggregate demand (AD), hence overall expansion in real gross domestic product with the attendant positive effect on stock market performance. The stock-oriented model, on the other hand, emphasizes the role of capital accounts in the determination of a country's exchange rate. In this theory, the exchange rate equates the demand and supply of financial assets (stock and bonds). Thus, expectation of future exchange rates affects the current price of financial markets (Aruori. 2011). From another perspective Brown & Yucel, (2002) aligning with the arbitrage price theory argued that a rise in real interest rate will reduce the present value (PV) of the future cash flow which consequently make stock returns to fall. As the real interest rate rises, capital flows increase, causing the domestic currency to appreciate and fall in stock returns, they argued. 2.1.3 Oil Price and Stock Price Oil price may impact stock performance through several channels such as uncertainty, fiscal, output and stock variation channels (Degiannakis, Filis & Arora, 2018). Oil price is susceptible to high volatility due to supply shocks and therefore, the risk of uncertainties occasioned by oil price volatility usually affect investors' portfolio, particularly, portfolio managers seeking to make optimal portfolio allocations (Arouri, 2011); cited in Salisu and Oloko (2015). Also, Uncertainty channels views explain that rising crude oil prices heightens uncertainty in the real economy (firms and households) due to its effect on inflation, consumption, and output (Brown & Yucel, 2002). For a firm, it tends to reduce the demand for irreversible investment and consequently, expected cash flow declines. On the household, increased uncertainty, resulting from higher cost of crude oil also increases the households' ability to save rather than consume. (Brown & Yucel, 2002). Given the above, the value of postponing investment and consumption decisions rises and hence, economic growth and stock market returns stifles. Oil prices also impact stock performance through a more direct channel— that is, the stock variation channel. The nexus suggests that stock returns are impacted by factors that can alter expected cash flows and discount rates. However, this depends on whether the firm is an oil user or oil producer. Ikemenogo, Eze Solomon (2024) American Interdisciplinary Journal of Business and Economics | 40 https://sadijournals.org/index.php/AIJBE Given that oil is a major production factor, any increase in oil price will result in an increased production cost (assuming a case of absence of substitution effect between production factors). This leads to reduced profit levels and future cash flows. For the oil producer, increase in crude oil prices results in increased profit margins and thus, increased future cash flows (Basher & Sadorsky, 2006). Furthermore, higher interest rate response by the monetary authority to an inflationary pressure from the rising oil prices also affect the discount rate; an important factor in stock price formulation (Basher, Haug & Sardosky, 2012). 2.2 Theoretical Review Theory-based framework from various ideas have been used in the past to support analyses of the impact of oil price volatility on stock markets. As a result, the theoretical foundation for this study comes from the Arbitrage Pricing Theory (APT). 2.2.1 Arbitrage Pricing Theory (APT) The Arbitrage Pricing Theory is a multi-factor asset pricing model which assumes that an asset's returns may be forecasted using a linear relationship between the assets expected return and several other macroeconomic variables that influence risk. Stephen Ross, an American economist, created this idea in 1976. The APT provides a multi-factor pricing model for securitized assets to analysts and investors. The APT provides analysts and investors with a multi-factor pricing model for securities that is based on the link between the projected return of a financial asset and its risk characteristics. The goal of APT is to determine the fair market price of a security that has been temporarily mispriced. APT is a more flexible and complicated alternative to the Capital Asset Pricing Model (CAPM). The theory allows investors and analysts to tailor their studies to their specific needs. Arbitrage is the practice of the simultaneous purchase and sale of an asset on different exchanges, taking advantage of slight pricing discrepancies to lock in a risk-free profit for the trade. APT provides traders with a model for calculating an asset's theoretical fair market value. Having determined that value, traders then look for slight deviations from the fair market price and trade accordingly. For example, if the APT pricing model determines the fair market value of a company's stock to be ₦50, but the market price drops to ₦45, the trader will buy the shares in the idea that additional market price action will rapidly "correct" the market price back to ₦50/share. Thus, this study uses the APT to link crude oil prices and other selected macroeconomic variables (such as the exchange rate, inflation rate, and interest rate) to stock market performance in Nigeria from 1981 to 2019. Its theoretical underpinning is derived from the APT. 2.2.2 Capital Asset Pricing Model (CAPM) The capital asset pricing model was pioneered by notable authors including Sharpe (1964), Umer (1965), Mossin (1960). The CAPM is a single factor model, quantifies the expected rates of return of an asset with level of market systematic risk. The CAPM has variously been lead among others by Chen (2003) and Kim et al. (a) 2017. It is a finance model that establishes a linear relationship between the required return on an investment and risk. CAPM evolved to measure this systematic risk. It is widely used throughout finance for pricing risky securities and generating expected returns for assets, given the risk of those asset and cost of capital. Algebraically, CAPM is presented as 𝑅𝑖 = 𝑅𝑓 + 𝛽𝑖(𝑅𝑚 − 𝑅𝑓) 2.1 𝑊ℎ𝑒𝑟𝑒: 𝑚 = 𝑚𝑎𝑟𝑘𝑒𝑡 𝑝𝑜𝑟𝑡𝑓𝑜𝑙𝑖𝑜, 𝑅𝑚 = 𝐸𝑥𝑝𝑒𝑐𝑡𝑒𝑑 𝑟𝑒𝑡𝑢𝑟𝑛 𝑜𝑛 𝑝𝑜𝑟𝑡𝑓𝑜𝑙𝑖𝑜, 𝑅𝑓 = 𝑅𝑖𝑠𝑘 𝑓𝑟𝑒𝑒 𝑟𝑒𝑡𝑢𝑟𝑛 𝑅𝑖 = 𝑅𝑒𝑡𝑢𝑟𝑛 𝑜𝑛 𝐴𝑠𝑠𝑒𝑡, 𝛽𝑖 = 𝐶𝑂𝑉(𝑅𝑖,𝑅𝑚) 𝛿𝑚 2 2.2 Ikemenogo, Eze Solomon (2024) American Interdisciplinary Journal of Business and Economics | 41 https://sadijournals.org/index.php/AIJBE Equation 2.2 is called the beta of the asset (𝑖) and (𝑚) is the variance of the market portfolio. For any portfolio < = (<1--- 0 and Β4 > or < 0 Exchange rates supposed to have positive relationship with stock market: due to globalization, businesses are affected either directly or indirectly by international trade activities. 3.2.2 Data Analysis Research has shown that most time series data poses unit root i.e. they are not stationary. Thus, research carried out with them is likely to be spurious or non-sense. A test of stationarity in time series data is very important because since the 1970s, macroeconomic aggregates in Nigeria have been fluctuating greatly. The consequence of using non-stationarity is so grave that well established models are breaking down as they continuously fail to predict outcomes. The problem according to Granger and Newbold, (1974) is that regression results on non- Ikemenogo, Eze Solomon (2024) American Interdisciplinary Journal of Business and Economics | 46 https://sadijournals.org/index.php/AIJBE stationarity series may, most times be “spurious or nonsensical” to the extent that a relationship would be accepted as existing between two variables as measured by their co-efficient of determination when in fact no relationships exist. Inference from non — stationary time series apart from being spurious, violate the classical econometric assumption, thus making the result unreliable for policy making. Also, pre-estimation test such as unit root tests and cointegration test were carried out while, post estimation tests such as stability test, normality test, auto correlation test etc. were also carried out to establish the consistency and reliability of the models adopted in this study.  Unit Root/Stationarity Test Adopting the Engel — Granger (1987) and Engel and Yule, (1987), we proceed to modelling a framework by first testing for stationarity to provide a more definitive answer to the non-stationarity, in each time series, the Dickey – Fuller (1979) regression is estimated as follows for unit root. ∆𝑌𝑡 = 𝜆𝑌𝑡−1 + 𝑉𝑡 3.5 If λ equals 0, 𝑌𝑡 is non-stationary, as a result 𝑌𝑡 and 𝑋𝑡 are not co-integrated. In order words, if λ is significantly different from 0, 𝑌𝑡and 𝑋𝑡 are found integrated individually. Given the inherent weakness of the unit root to distinguish between null and the alternative hypotheses, it is desirable that the augmented Dickey — Fuller (ADF), 1981 test be applied. To be co-integrated; both 𝑌𝑡 and 𝑋𝑡 must have the same order of integration (Engel and Granger, 1987, and Granger, 1986). The ADF regression is specified as follows: ∆𝑌𝑡 = 𝛼 + 𝛽1𝑡 + 𝛿𝑋𝑡−1 + ∑ 𝛽𝑖∆𝑋𝑡−𝑖 𝑚 𝑡=1 + 휀𝑡 3.6 ∆ is the first difference operator, 6.466.486.406.506.526.566.546.426.44 Akai ke Infor mati on Criteri a (top 20 models) is the new random error term, M is the optimum number of lags needed to obtain “white noise’. This is approximated when the DW values approaches 2.0 numerically. The null hypothesis of non-stationarity is rejected if the estimated ADF statistic is found to be larger in absolute term or more negative than its critical values at 1 or 5 percent level of significance.  Concept of Co-integration Co-integration among the variables is used to determine the existence of a long run equilibrium relationship between the variables. The concept of co-integration (Granger, 1986, Mill 1990) creates the link between integrated processes and the concept of steady state equilibrium. The idea behind co-integration is that ‘although two different series may not themselves be stationary, some linear combination of them may be indeed stationary with the generalization to more than two series” (Komolafe, 1996). Economic variables are inherently non- stationary and thus, could meander without any tendency to return to equilibrium in the long run. Implicit in the co-integration theory is the fact that there exists a linear combination of these non-stationary variables that is stationary. The traditional approach to the modelling of short-run disequilibrium is the partial adjustment method. However, an extension of this in the co-integration technique is the Error correction mechanism (ECM) (Granger and Newbold, 1977). The original co-integration relationship is specified as follows: 𝑌𝑡 = 𝛽0 + 𝛽1𝑋𝑡 + 𝜇𝑡 3.7 Analysing the long-run behaviour of 𝑌𝑡 implies investigating co-integrating relationship in (1). If µt is stationary then, the 1(1) variables in 𝑋𝑡 may be thought of as capturing the long run component of 𝑌𝑡 while 휀𝑡 captures the short run or temporary movements.  Error correction technique If the 𝑌𝑡 and 𝑋𝑡 are found to be co-integrated, then there must exist an associated Error Correction Model (ECM), according to Engel and Granger (1987). The usual ECM may take the following form: Ikemenogo, Eze Solomon (2024) American Interdisciplinary Journal of Business and Economics | 47 https://sadijournals.org/index.php/AIJBE ∆𝑌𝑡 = 𝛼0 + 𝛽1𝑡 ∑ ∆𝑋𝑡−1 𝑚 𝑡=1 + 𝛿𝑖 ∑ ∆𝑌𝑡=1 𝑚 𝑡=1 + 𝛾𝑡휀𝑡−1 + 𝜑 3.8 Where, ∆ denotes first difference operators, 휀𝑡−1 is the error correction term, m is the number of lags necessary to obtain “white noise” and 𝜑𝑡 is another random disturbance term. If /δ/ is significantly different from zero, then 𝑌𝑡 and 𝑋𝑡 will have longer run relationship. The (ECM) error correction term (휀𝑡−1) depicts the extent of disequilibrium between 𝑌𝑡 and 𝑋𝑡. the ECM, reveals further that the change in 𝑌𝑡 not only depend on lagged changes in 𝑋𝑡 but also on its own lagged changes. The estimate of the parameters of the ECM are generally consistent and efficient (Hendry and Richard, 1983). Inference about the long run Granger causality can be drawn from the ECM model. The presence of co-integration will indicate at least, unidirectional long run causality from ∆𝑋𝑡−1, if statistically significant will indicate a short run causality from ∆𝑋𝑡−1 to ∆𝑌𝑡−1. The statistically significant non-Zero co-efficient of ∆𝑌𝑡−1will indicate feedback to ∆𝑌𝑡 from its own lagged values. It may be noted that even in the absence of co-integration, the error correction model may be estimated to detect if there is any short run granger causality. 3.2.3 Nature and Sources of Data The data used in the study are collected from various publications of the Central Bank of Nigeria and (CBN), World Development Indicator (WDI), National Bureau of statistic (NBS). Specifically, the data used are time series data, which includes the exchange rate (EXCH), interest rate (INTR), inflation (INFL), oil price (OILP) and Real GDP(RGDP). 3.2.4 Criteria for Model Evaluation To analyse the model, we employed the economic criteria which is used to measure the sign and size of the parameters in the model. Statistical Criteria: This includes the T-statistic, F-statistic, and 𝑅2 Co-efficient of determination (𝑹𝟐) Coefficient of determination also known as R square (𝑅2) or goodness of fit tells the proportion of the total variable in the dependent variable y that is explained by the regression line or the explanatory variables x. in a single / simple regression model, it is the square of the correlation coefficient in a simple regression model. The value of the 𝑅2 lies between 0 and 1 i.e. 0< 𝑅2<1 when the 𝑅2 = 0, it means the explanatory variable do not explain the dependent variable and when the 𝑅2 = 1, it means the model is best fit. If the 𝑅2 is multiplied by 100, then it shows the percentage of total variation in y the dependent variable that is explained by variation / changes in x. the closer the 𝑅2 is to one. The stronger is the explanatory power of the estimated regression line, and thus the closer are the observation to the line. If 𝑅2 = 0.56 it means that 56% of total variation in y is explained by the regression line / changes in x and the other 44 remains unexplained by x. Test of individual statistics of the slope coefficient (T. Test) This tests the individual significance of the co-efficient, to do this, we test the null hypothesis that bi= 0 against the alternative hypothesis that bi. ≠0 in employing the t test, we compare the computed t-value with the value read from the student’s t table at given level of significance, (α) and n-k degree of freedom/ if the absolute value of the computed t value is greater than the absolute value of the theoretical, t value, we reject the null hypothesis (H0) at the given level of significance i.e. if 𝑡𝑐𝑎𝑙>𝑡𝑡𝑎𝑏, reject H0; Accept H1 𝑡𝑐𝑎𝑙<𝑡𝑡𝑎𝑏 , Accept H0, reject H1. Alternatively, the rule of thumb can be used, which state that if the 𝑡𝑐𝑎𝑙 is greater than 2 at the 5 percent level of significance, we reject the null hypothesis and conclude that the parameter is statistically significant in explaining the dependent variables. Ikemenogo, Eze Solomon (2024) American Interdisciplinary Journal of Business and Economics | 48 https://sadijournals.org/index.php/AIJBE The F Statistic (Test for overall significance of the model) This is used to test for the statistical significance of the entire slope coefficient jointly on the dependent variable using a level of significance. The F statistic is done by comparing the 𝐹𝑐𝑎𝑙with the F tabulated. When the 𝐹𝑐𝑎𝑙 is greater than the 𝐹𝑡𝑎𝑏 we reject the null hypothesis and conclude that the entire variable put together is statistically significant in explaining the dependent variable. Econometric Criteria This includes the test for serial correlation. D.W Statistic Durbin – Watson Test It is the most popular test for serial/autocorrelation. The Durbin-Watson statistic is used to test the presence of serial correlation in a model, to determine whether there is serial correlation in the model, we compare the Durbin Watson value from the model to the Durbin-Watson critical value at 5%. If the value of Durbin Watson lies between the upper limit and four minus upper limits (i.e. u