Multiple Regression Analysis – Term Paper

Statistic definitions

Beta coefficient of stock is used to measure market risk. beta coefficient is calculated with a graph, with one variable on the Y-axis and the other two on the X-axis. The slope of the characteristic or regression line can indicate the level of business friendliness. The regression line is used to evaluate the relationship of the variables. The b₀ intercept on the line can be used to determine the basis of identifying the results of the analysis but does not have any impact on the results on multiple regression. Therefore, it cannot be useful in interpreting the results. In this case, the first important value is b₁ through to the last value of b₆.

Multiple regression analysis example

The results explained

In the survey, various states and metropolitans are graded regarding their overall business friendliness and the friendliness to the local labor regulations. The approach of dominant analysis is also employed in determining the relationship of the business friendliness, change in revenue and economic change. The performed calculations reveal these variables have either a positive or a negative correlation. The concept of regression analysis was also used to determine the positive changes along various policy measures impacts perception of overall friendliness. The economy should be stable and the revenue should relatively decline to enhance business relations.

Econometric analysis enabled the calibration of parameters of the model and the estimate statistical relationships of the provided variables. In this context, the business friendliness is compared to changes in the state economy and change in revenue. A decrease in the state economy result in decline in business friendliness, hence the two portray a positive correlation. A positive correlation refers to the tandem movement of variables that is if one variable increases the other dependent variable decreases as well and vice versa. On the other hand, a decrease in revenue results in an increase in business friendliness, this relationship indicates a negative correlation of variables. A negative correlation occurs when the relationship between two variables is not dependent on each other, and an increase in one results in a decrease in the other variable, and vice versa.

Multiple regression analysis for dummies

Interpretation of the results

Rejection rule = We use the rejection rule for a two-tailed tests. Considering the confident interval is 0.05, the common Z test for this interval is -1.960 and +1.960 for the t statistic. Therefore, in regression 1, there is only one area with greater business friendliness, which is the Asian, which also applies to the case in regression 2. This means that there is high business friendliness in the Asian niche. The several metropolitans have labeled into x₁ through to x₆, for regression 1, and x₁ to x₇, for regression 2. The Asian metropolitan depicts more business friendliness, and can be induced further when the x₇ aspect is generally introduced. From the logic, there is lower projection of business friendliness, in the second regression, whose only difference is the introduction of government support for small businesses.

Impact of the introduction X7

The introduction of Government support for small businesses affects every X, in the dataset, with which some are affected positively, while other are affected negatively. This can be defined by the closeness that it moves to the Z test score. In this case, we defined it to be a range between -/+ 1.960. The Asian Metropolitan is not affected so much because it remains within the range. However, the other metropolitans, represented by x1 through x4, and x6, are all moved closer but not within the Z test range. This means that there is a general slight improvement on other metropolitans on creating an average set of business friendliness. However, on the general view of friendliness, there is a general drop of the average of all the Xs, because it affects the major business friendly niche, which is the Asian metropolitan.