Regression Analysis and Estimate Demand Curve - Essay Example

Regression Analysis and Estimate Demand Curve Data collection is the first step of estimating demand using regression analysis where data collected includes the prices that prevailed over a certain period against the corresponding amount of sales that were made within that period (Samuelson and Stephen, 133). The points are then plotted in a scatter diagram to show the relation between the sales and price giving a line that touches most of the points and this line is translated as Q= α + βP being the linear demand equation.

Coefficient α and β represent the constant term and the slope of the demand equation respectively, forming a linear equation and is preceded by the generation of numerical values of α and β (Samuelson and Stephen, 137).Ordinary Least Squares regression method is used for the determination of the coefficient values. Arbitrary values of α and β are taken and the sales values at the respective prices generated then a difference between these values and the amount of actual sales is calculated.

This gives estimation errors where the OLS regression method squares these errors and adds them to find the sum of squared errors (SSE) measuring the accuracy of the equation. Through calculus and statistical formulas, the values for α and β are calculated giving the demand equation. However, demand is affected by different variables apart from price leading to the need to include other variables in the equation and through OLS regression; the coefficients for the other factors affecting demand are calculated creating an equation with a number of independent variables affecting demand.

The t statistic is the value of a coefficient divided by its standard deviation. The statistical tests t increases the confidence in the results because it shows if a certain coefficient of demand determinant has explanatory power (Samuelson and Stephen, 145). The confidence in the results is increased through statistical f tests because it allows the statistical significance of the regression equation (Samuelson and Stephen, 143). The statistical tests R2 (coefficient of determination) measures the proportion of variation in the dependent variable, which is the goodness of fit (Samuelson and Stephen, 141).

When R2 depicts a high goodness of fit, it increases the confidence in the results. Work CitedSamuelson, William, and Stephen G. Marks. Managerial Economics. Hoboken, NJ: John Wiley and Sons, 2012. Print.