Importance of Econometrics Modelling - Essay Example

ECONOMETRICS MODELLING EXECUTIVE SUMMARY The key objective of each and every organization is to maximize the income generated while minimizing on the costs. Sales made by the company forms the basis of income generation-the more the sales, the more the income is expected. 3M Company operates as a diversified technology company worldwide and as such a number of variables such as USA nominal GDP, USA unemployment rate and the PMI index are expected to affect the sales made by the company. This report examines how the company can predict the sales it can make without having to go into deep details. The company can estimate the sales to be using a regression model. The sales depend on the variables mentioned such as the USA nominal GDP, USA unemployment rate and the PMI index. The report provides a correlation matrix and regression model with the data on the sales being regressed and the output analysed. The correlation matrix shows that the variable for sales is significantly correlated with the USA GDP nominal with a positive Pearson coefficient of .957 (which shows a very strong positive correlation). The other two variables are also correlated with the dependent variable (with PMI index being negatively correlated with the dependent variable while USA unemployment rate having a positive correlation) they are however not significant at 5% significance level. The results from the regression model show that the explanatory variables have a lot of impact on the dependent variable (Sales). The study established that approximately 95.91% of variation in the dependent variable (sales) is explained by the three explanatory variables (USA nominal GDP, USA unemployment rate and the PMI index) in the model. INTRODUCTION According to Creswell (2003), regression analysis is a statistical tool that is usually utilized by many researchers to investigate the relationship that exists if any between two or more variables. The aim of the investigation is to find a causal effect relationship. Using the data, the researcher can assess the statistical significance of the relationships that have been estimated. The level of confidence that is to be established is that the estimated relationship is close to the actual relationship. Regression analysis has been in use for many years and it has increasingly been applied in various disciplines (Jankowicz, 2005). This study uses regression analysis to establish the relationship that exists between the actual sales made by 3M Company and the independent variables. Using a regression model, the company will find a viable way of determining its sales pegged on the explanatory (independent) variables given. DATA The data that are used in this paper have been downloaded from Bloomberg for 3M Company that operates as a diversified technology company worldwide. Bloomberg is a premier site for business and financial market news. It delivers world economic news, stock futures, stock quotes, & personal finance advice. The data illustrate the sales made against the social-economic variables (USA nominal GDP, USA unemployment rate and the PMI index) from 31st December 2004 to 31st December 2013. THE MODEL In order to explore the relationship between the dependent variable (sales) and the explanatory variables (USA nominal GDP, USA unemployment rate and the PMI index), we used a multiple regression model. Before fitting any trend and or seasonal dummies to the model, we must include an intercept, because we are obviously not starting at zero. The model used is thus as shown below; Where, is the coefficient for the constant term, is the coefficient for the USA nominal GDP, is the coefficient for the PMI index, is the coefficient for the USA unemployment rate and lastly, is the error term. DESCRIPTIVE STATISTICS In this section, we present the descriptive statistics for the variables which include the mean, median, maximum, minimum, standard deviation, skewness, kurtosis and the Jarque-Bera tests. The standard deviations for the variables show that the variables are not sparsely distributed. On the other hand, the Jarque-Bera test shows that out of the four variables only one variable (PMI index) does not possess the qualities of a normal distribution (that is, PMI index does not follow a normal distribution) while the rest of the variables follow a normal distribution. PMI_INDEX SALES_USA_ONLY__USDM USA_GDP_NOMINAL__USDBN USA_UNEMPLOYMENT_RATE___  Mean  51.86000  9259.000  14625.30  6.943000  Median  54.10000  9083.000  14598.10  6.920000  Maximum  57.50000  11151.00  16768.10  9.930000  Minimum  33.10000  7878.000  12274.90  4.430000  Std. Dev.  7.212520  1026.468  1350.054  2.015264  Skewness -1.910644  0.557870 -0.134113  0.192082  Kurtosis  5.769838  2.303292  2.367243  1.626583  Jarque-Bera  9.280933  0.720950  0.196803  0.847440  Probability  0.009653  0.697345  0.906285  0.654607  Sum  518.6000  92590.00  146253.0  69.43000  Sum Sq. Dev.  468.1840  9482732.  16403811  36.55161  Observations  10  10  10  10 MULTIPLE REGRESSION ANALYSIS The model to be estimated is; From the regression table below, we observe that , , and . The model can thus e estimated as follows; Dependent Variable: SALES_USA_ONLY__USDM Method: Least Squares Date: 12/30/14 Time: 18:44 Sample: 2004 2013 Included observations: 10 Variable Coefficient Std. Error t-Statistic Prob.   Intercept -2645.163 1212.009 -2.182460 0.0718 USA_GDP_NOMINAL__USDBN 0.823360 0.073581 11.18983 0.0000 PMI_INDEX 13.50073 12.14417 1.111705 0.3088 USA_UNEMPLOYMENT_RATE -120.6780 49.65696 -2.430233 0.0511 R-squared 0.959067     Mean dependent var 9259.000 Adjusted R-squared 0.938601     S.D. dependent var 1026.468 S.E. of regression 254.3472     Akaike info criterion 14.20445 Sum squared resid 388155.0     Schwarz criterion 14.32549 Log likelihood -67.02226     Hannan-Quinn criter. 14.07168 F-statistic 46.86054     Durbin-Watson stat 1.713306 Prob(F-statistic) 0.000148 From the regression table, we also observe that the p-value for the F-statistic is 0.0001 Read More