In this chapter is going to describe the research design, I will be using to tackle the research problems stated in chapter one.
3.1 RESEARCH DESIGN
This study aims at showing the impact of the mobile telecommunication sector on economic growth in Nigeria; It assumed that the major benefits of the Nigerian mobile telecommunication sector towards the growth of the economy is the attraction of foreign direct investment. In this case, we will be evaluating the effect foreign direct investments in the telecommunication sector, and in turn the effect of telecommunication sector on foreign direct investment. The information for evaluating this effects gotten from historical data documented in the World Bank websites and CBN economic and financial reviews from the past to present in the country.
3.2 DATA COLLECTION AND SOURCES
In other to accomplish the stated objectives of this study, secondary data were sourced and used. Dataset identifying with aggregate Foreign Direct Investment (FDI) and the Gross Domestic Product (GDP) sourced from Central Bank of Nigeria (CBN) Statistical Bulletin and the World Development Indicator of 2014 from World Bank. Telecommunication data collected from trading economics website, the data of Nigerian FDI gotten from the same website was also used to compare the FDI data gotten from CBN bulletin.
The period covered by this study is 2002-2014 and the decision of period covered was made based on the advancement time of telecom industry in Nigeria.
3.3 METHOD OF DATA ANALYSIS
First of all this research will carry out a Unit Root test, which is a test done on a time series variables to test whether they are stationary or non-stationary. Testing the Stationarity matters for so many reasons:
– The Stationary or Non-Stationary can strongly influence its properties and behaviours, thereby given inaccurate results.
– If the variables in the regression model are not stationary, then it can be demonstrated that the standard assumptions for asymptotic analysis will not be valid. As such, the usual “t-ratios "will not follow the t-distribution, so we cannot truly attempt hypothesis tests about the regression parameters.
-It will result to spurious regressions. Because if two unrelated variables are trending over time the regressions of one could give high R^2 despite the fact they are not related.
These are some of the reason testing for stationarity is very important, and to that effect, the study will adopt a Philip-perron test that is an important Unit root test that will use with Eviews software. There is another test such as:
-Dickey-Fuller test
-Augmented Dickey-Fuller test
-KPSS test
-Zivot-Andrews test
Though to narrate other common unit root test such as the Dickey-fuller test and also its adjustments known as Augmented Dickey Fuller test is important before we get to Philip-Perron test. While a simple AR (1) model is:
Y_(t = ) 〖Py〗_(t-1)+ U_t
Where Y_(t )the variable of interest is, is the time index, P is a coefficient, and U_t is the error term. Also a unit root is present if P=1, the model would be non-stationary in this case.
The Dickey-Fuller test developed in 1979 is simply the result of an Algebra:
〖∆Y〗_(t = ) 〖αy〗_(t-1)+(constant and time trend)+U_t
Where α = P-1
It involves fitting regression model by ordinary least square (OLS), but the serial correlation will present a problem. These where the Augmented dickey fuller test comes in because it includes lag of the first difference of 〖∆Y〗_(t )regressors in the test equation. The Augmented Dickey Fuller test is an Augmented version of Dickey Fuller, wh ich is used for a more larger and complicated sets of time series models, but the U_t might be heteroskedastic.
As an Economist while researching we are tasked with four major things, describing and summarizing macroeconomics data, make forecasts with this data, quantify what we do or do not comprehend about the genuine structure of the macroeconomy, and finally give our advice or opinion on could be. To achieve all that we employ the Vector autoregression (VAR). A VAR is an n-equation, the ‘n’ variable linear model in which each variable explained by its lag values, plus current and past values of the remaining n-1 variables. VAR comes in three forms, reduced form, recursive and structural. The decreased form VAR expresses each variable as a linear function or capacity of its own past values or qualities, the past estimations of all other variables considered, and serially uncorrelated error term. A recursive VAR constructs the error terms in each regression equation to be uncorrelated with the errors in preceding equations. While Structural VAR uses economic theory to sort out the contemporaneous link between variables.
There three (3) VAR analysis normally was done by researchers
– Granger-Causality Tests,
-Impulse response functions
-Forecast error variance disintegrations
The most interesting to researchers is the impulse response function, which will be analysed in the next chapter. The impulse responses trace out the reaction of present and future estimations of each of the variables to one unit increment in the present value of one of the VAR errors, accepted that this error returns to zero in subsequent periods and that every single errors are equal to zero (Stuck & Wathson, 2001).
3.4 MODEL SPECIFICATION
In other to attain the desired result to our research question, the methodology includes evaluating an econometric model in which the relationship between FDI, TEL and economic growth in Nigeria is tested for. Therefore, we adopt a Solow growth model formulated by Robert Solow. The Solow model believes that a managed rise in capital investment increases the growth rate only brief in light of the fact that the ratio of capital to labour goes up. On the other hand, the marginal product of extra units of capital may decline (there are diminishing returns) and thus an economy goes back to a long-term growth way, with real GDP growing in proportion to the growth of the workforce in addition with factor to reflect improving productivity. The model shows how by investing on existing technology, you could increase growth.
Solow growth model has a Short run and Long run implication:
In the short run, growth is determined by moving to the new steady state that is made just from the adjustment in the capital investment, labour force growth and depreciation rate. The adjustment in the capital investment is from the adjustment in the savings rate. While in the Long run, growth is achievable only through technological progress. Therefore, some important statistical test such as Impulse response function and Johansen cointegration test will be used to check if there is a positive impact between the variables in the long run and also on a short run. E-Views was employed to analyse the relationship between two variable foreign direct investment (FDI) and investment in telecommunication industry (TEL), the third variable Gross domestic product (GDP) serves as the control variable. It is interesting to note that greater percentage of the telecommunication investment in Nigeria were financed through FDI. Hence, it was assumed that significant relationship exists between the influx of FDI and telecommunication growth in Nigeria.
3.5 DESCRIPTION OF VARIABLE
In other to analyse the impact of the telecommunication sector on economic growth in Nigeria, three variables GDP, FDI and TEL were considered. GDP was used as a controlled variable while checking if FDI had any significant effect on TEL, which also goes to check if TEL had any significant effect on FDI. The statistical formulation of the model is therefore presented as follow:
FDIt =β1 + ∑_(i=1)^n▒〖α_1 TEL_(t-i) 〗 + ∑_(i=1)^(n )▒〖μ_1 FDI_(t-i) 〗 +∑_(i=1)^(n )▒〖∞_1 GDP_(t-i) 〗 + et …. (1)
TELt =β2 + ∑_(i=1)^n▒〖α_2 TEL_(t-i) 〗+ ∑_(i=1)^(n )▒〖μ_2 FDI_(t-i) 〗 +∑_(i=1)^(n )▒〖∞_2 GDP_(t-i) 〗 + et…. (2)
Where:
FDI = Foreign Direct Investment in Telecom Industry.
TEL= Investment in Telecommunications Industry.
GDP= Gross domestic products.
Also, n denotes the numbers of lag that were determined by the information criterions, β1-2, α1-2, ∞1-2 and ρ1-2 are parameters for estimation, and et and ut are residual terms. To detect a causal relationship between the variables we used the VAR Granger causality technique. There were several test conducted to derive results for the purpose of this study using an Eviews software. Such test as Philip-perron test which is under the Unit root test, and Johannsen Cointegration test together with the General Impulse Response Function were carried out to analyse and check if there was a significant impact of FDI, Tel and GDP at various time duration or intervals. It is also very important to note that as much as foreign direct investment FDI is financing the Nigerian telecommunication sector, the telecom sector also returns the favour by attracting more investments into the country.
3.5.1 FOREIGN DIRECT INVESTMENT
Foreign direct investment (FDI) is an investment that involves the injection of foreign funds into a company or an enterprise; that operated in another country. In other for an investor to have a say or voting right in a foreign company, he or she must own shares of over or equal to 10%. Otherwise, anything lower than 10% will be termed as portfolio investment, not FDI. Foreign direct investment (FDI) could be inward or outward. An inward FDI involves the investment of foreign capitals into local resources, normally propelled by the lowering of tax charges, interest rates and grants. The outward FDI involves the direct investment abroad, and it is carried out or supported by the government.
In the case of this research paper, we are referring to the inward FDI, meaning the inward investment of funds in the mobile telecommunication industry and how it affects the economic growth of the country. Therefore, we shall be establishing the determinant of FDI, based on assumption as:
FDI= β1 + β2 TEL + β3 GDP + µ………………….. (3)
Where:
FDI= Foreign Direct Investment in Telecom Industry.
Tel= Investment in Telecommunication industry
GDP= Gross domestic products.
µ= Error term.