Topic_03_Elasticity_Estimation (1).pptm

On the other hand if the regression explains none of

Info icon This preview shows pages 31–39. Sign up to view the full content.

View Full Document Right Arrow Icon
On the other hand if the regression explains none of the variation in the dependent variable, then R 2 =0.
Image of page 31

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full Document Right Arrow Icon
Goodness of Fit and R 2 Statistics (a) R 2 = 0.98 (b) R 2 = 0.54 R 2 = 0.54 on the right panel shows that 46% of the variation in demand is due to random errors. 32
Image of page 32
Do We Have the Right Functional Form? – Advertising and Demand (a) Linear (b) Quadratic Using a linear regression would be a mistake in this case. The actual relationship is more like a quadratic function. 33
Image of page 33

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full Document Right Arrow Icon
3.3 Statistical Significance of Estimated Coefficients How close the estimated coefficients are to their respective true values ? If the standard errors ( ) of the estimated values are low, then these values are good indicators of true values. An estimation method is unbiased if it produces estimated coefficients that are equal to true values on average. OLS is unbiased method. 34
Image of page 34
Confidence Interval/Hypothesis Testing A manager may want to test if the price of rival product ( p r ) affects it demand ( y ). The estimated equation is: = + p r The estimated coefficients (and ) are less likely equal to their true values ( β 0 and β 1 ). In this case, we can construct a 95% (or any other) confidence interval such that the true value lies within this interval 95% of times. 35
Image of page 35

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full Document Right Arrow Icon
Hypothesis Testing The null hypothesis that rival’s coefficient is zero. Null hypothesis H 0 : = 0 Alternative H 1 : ≠ 0 If 95% confidence interval for includes zero can be tested by using t-statistic = ( - 0)/ . If the absolute value of t-statistics is > critical value (which is close to 2 for two-tailed 95% level), then we reject null hypothesis. 36
Image of page 36
Hypothesis Testing That is, 95% confidence level (CI) does not include zero. There is 95% probability that rival’s price affects the product’s demand. Or simply that the explanatory variable is statistically significant (at 5% significant level or 95% confidence level) . 95% CI = ± t 95%, n-k-1 . If n= # of observations, k = # of explanatory variables, then n-k-1 = degree of freedom. 37
Image of page 37

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full Document Right Arrow Icon
3.4 Regression Specification It is important to correctly select the explanatory variables and distinguish correlation and causation . Correlation may NOT necessarily imply causation. For example, during the summer time incidence of both gasoline demand and sun burn are high, and so there is a strong correlation between them. Does that mean demand of gasoline causes sun burn? 38
Image of page 38
Causation and Omitted Variables Obviously, NO ! There is a third factor in this case that is sunny days that is affecting both. Sometimes your analysis may omit some of the explanatory variables (not taken into consideration). In that case the resulting coefficient estimates may not be reliable. 39
Image of page 39
This is the end of the preview. Sign up to access the rest of the document.

{[ snackBarMessage ]}

What students are saying

  • Left Quote Icon

    As a current student on this bumpy collegiate pathway, I stumbled upon Course Hero, where I can find study resources for nearly all my courses, get online help from tutors 24/7, and even share my old projects, papers, and lecture notes with other students.

    Student Picture

    Kiran Temple University Fox School of Business ‘17, Course Hero Intern

  • Left Quote Icon

    I cannot even describe how much Course Hero helped me this summer. It’s truly become something I can always rely on and help me. In the end, I was not only able to survive summer classes, but I was able to thrive thanks to Course Hero.

    Student Picture

    Dana University of Pennsylvania ‘17, Course Hero Intern

  • Left Quote Icon

    The ability to access any university’s resources through Course Hero proved invaluable in my case. I was behind on Tulane coursework and actually used UCLA’s materials to help me move forward and get everything together on time.

    Student Picture

    Jill Tulane University ‘16, Course Hero Intern