Question 1 (answer)
Ho is a null hypothesis since it shows there is no relationship amongst the variables and cannot be proven mathematically while H1 is an alternative hypothesis since β≠0 but it is a figure above zero, β=10.85.
Ho:β2=0 is a null hypothesis since the value of β is greater than 0. Ho:β2 is an alternative hypothesis because its value is not zero as indicated. There is a positive relationship which can be proven. β2=8.58
Ho: β3 >0
Ho:β3=0 is a null hypothesis since its value is a non-zero figure and cannot be proven by mathematical means on the other hand H0:β3>0 is an alternative hypothesis which indicates a negative relationship between the variables.
Ho:β7=0 is a null hypothesis as it can be clearly seen that its value is a non-zero figure ,it cannot be proven through mathematical means. Ho:β7=0.40 is an alternative hypothesis showing a negative relationship between the variables. It can be proven mathematically.
There is an apparent contradiction between results in (i) and (ii) above in the sense that β2 is indicated to be equivalent to 8.58 in (i) and it is further indicated to be equivalent to a zero (0) by a null hypothesis in (ii).(ii) is a contradiction since the value of β2 has already been given as 8.58.
Basing my argument on the empirical model and the hypothesis tested above in question one, The US$2 billion would be effectively spent on bailing out the banking sector rather than funding secondary education. The banking sector contributes substantially to the economy through advancing credit to individual investors and corporations especially in times of financial crisis (Kemerer, 1987: 120). Failure to bail out the banking sector means a drop in the ratio of Private Credit by Deposit Money Banks and Other Financial Institutions to GDP, from the current figure of 0.52 to an estimated 0.38.In addition, maintaining a strong and vibrant banking sector is critical if the country is to achieve the targeted growth in its per capita GDP over the medium term .By channeling the US 2 billion to the banking sector, the government takes over the bad debts reported by commercial banks and avails credit to entities in need of cash for investment purposes. In the attempt to raise the per capita GDP, an economy should focus on policies that will boost the economy and avoid those that will lead to inflationary pressures and other economic difficulties. Such policies that may lead to economic downturn include increasing the government expenditure such as the proposed idea of funding the secondary education using the available US$2 billion (William, 1991: 81). Although this move will boost the secondary school enrolment rate (expressed as a percentage of the secondary school age population) by about 6%, from the current figure of 55% to an estimated 61%, it has economic implications and may not lead to increase in per capita GDP over the next 15 years. I would therefore advice the FINANCE MINISTER to bail out the banking sector with the available US$2 billion taking into account this is a medium term strategy that is expected to increase per capita GDP over the next 15 years.
- a. dlypci = b1 + b2lypc90i + b3lsecedi + b4govgdpi + b5openi + b6infli + b7crediti + ui
According to the linearity assumption, the relationship between the predictors and the outcome variable should be linear. Thus this statistical assumption is invalid as can be indicated above by the linear equation above. The sum of the proctors does not lead to the outcome indicated.
- According to homoscedasticity assumption, there ought to be consistency in error variance. Our empirical model shows is valid in the sense that the errors are constant throughout. Significance level is either 0.05 or 0.1.
- Normality assumption stipulates that the distribution of errors should be normally .As can be indicated in the stata, errors are not normally distributed and the coefficients are also not independently distributed in all fifty countries. This shows that the empirical model is invalid in terms of normality assumption.
Outlier refers to an observation which has a big residual or rather an observation that has a value of (dependent variable) that are unusual according to the given values on the predictor variables (Lawrence, 2003: 240). An outlier is very sensitive in an empirical model as it is an indication of errors that may take some time to detect. Specifically outlier could show sample oddness or could refer to an error related to data entry and other countless problems.
Creating a 0-1 dummy variable for each outlier you, and re-estimating the model. Using the new specification gives the results below;
Re-specification and re-estimation of the model does not alter my advice to the minister for finance. This is in relation to how proposal should be undertaken, either bailing the banking sector or funding secondary education in an attempt to increase the per capita GDP over the next fifteen years (Myers, 2003: 45). I still recommend bailing out the banking sector since even after creating a 0-1 dummy variable for outlier I identified in the data I used to estimate the model, the model still disobeys the normality assumption meaning that the model still has several errors that have not been identified (Randall, 2003:24). This justifies my decision to still go for the bailing of the banking sector with the US$2 billion I had recommended earlier.