Inferential Statistical Analysis: Evaluating Empirical Results

This assignment provides a review of inferential statistical analysis, focusing on the evaluation of empirical results and the suitability of different statistical estimation and hypothesis testing procedures. It includes the analysis of an OLS regression model examining the impact of oil prices, interest rates, and inflation on US real GDP growth, discussing the statistical significance of parameters, interpreting the sign and magnitude of estimates, and assessing the overall fit of the model. Additionally, it covers the analysis of Mann-Whitney tests comparing productivity between different groups of employees (male vs. female, postgraduate vs. undergraduate, trained vs. non-trained), evaluating the statistical significance of the coefficients and the overall suitability of the regression model. The review also touches upon the theoretical underpinnings of the results and suggests improvements to the models based on the findings.

REVIEW OF QUANTITATIVE LITERATURE REVIEW
STUDENT ID:
[Pick the date]

Question 1
Part A
 Statistical significance of the given parameters
Assuming 5% level of significance
The variables would be considered as statistically significant when the p value is lower than
level of significance (Lehman and Romano, 2016). It can be seen from the above that the p
value for both OIL and INTERESTRATE is lower than significance level (0.05) and thus it
can be said that both the variables are statistically significant. The p value for INFLATION is
higher than significance level and thereby it is not statistically significant. It can be concluded
that in the current model of computing the REALGDP the INFLATION variables are
insignificant.
 Sign and magnitude of the estimates
OIL slope coefficient = -0.037 (Negative sign), It implies that one unit increase in OIL would
reduce the REAL GDP by 0.037 units.
INTERESTRATE slope coefficient = -0.012 (Negative sign), It implies that one unit increase
in INTERESTRATE would decrease the REAL GDP by 0.012 units.
INFLATION slope coefficient = -0.004 (Negative sign), It implies that one unit increase in
INFLATION would reduce the REAL GDP by 0.004 units.
 Overall utility of model
The adjusted R2 of the model = 58%
The value comes out to be higher than 50% which implies that the model is a good fit.
However, there is one variable in the form of inflation which ought to be removed and other
more relevant independent variables should be substituted (Harmon, 2016).
2

Part B
Yes! It can be said that the result is in line with the prediction of the principle because
increase in oil price, inflation and interest rate would negatively affect the real GDP. This is
because increase oil price leads to inflation owing to which the interest rate would be
increased. The higher interest rate would have adverse impact on investment and
consumption related expenditure leading to lower GDP growth (Taylor and Cihon, 2017).
Question 4
Part A
Regression result
 Sign and magnitude of the coefficient
Age slope coefficient = 0.035 (Positive sign), It implies that unit increase in the age of
would increase the odds of promotion by3.50%.
Experience slope coefficient = 0.148 (Positive sign), It implies that unit increase in the
experience of people would increase the odds of promotion by 14.80%.
Sex slope coefficient = -0.986 (Negative sign), Considering male as the baseline, it can be
said that for a female the chances of promotion decreases by 18.66% as compared with the
Thus, it can be concluded that age and experience positively affect the promotion of the
people in the past 24 months whereas Sex/Gender negatively affects the promotion of the
people.
 Statistical significance of the coefficient
3

Assuming 5% level of significance
The variables would be termed as statistically insignificant for the model when the p value is
higher than level of significance. It can be seen from the above that the p value for all the
variables is higher than significance level (0.05) and thus it can be said that all the given
variables are statistically insignificant (Harmon, 2016).
Part B
It can be said that none of the variables is statistically significant and therefore, it can be said
that this regression model is not a good fit for analyzing the promotion of people. It is
required that relevant independent variables should be included into the model instead of the
current variables which are not satisfactory (Medhi, 2016).
4

References
Lehman, L. E. and Romano, P. J. (2016) Testing Statistical Hypotheses. 3rd ed. Berlin :
Springer Science & Business Media.
Medhi, J. (2016) Statistical Methods: An Introductory Text. 4th ed. Sydney: New Age
International.
Taylor, K. J. and Cihon, C. (2017) Statistical Techniques for Data Analysis. 2nd ed.
Melbourne: CRC Press.
Harmon, M. (2016) Hypothesis Testing in Excel - The Excel Statistical Master. 7th ed.
Florida: Mark Harmon.
5