Jennifer Back Econometrics & Forecasting Regression Presentation
2 OBESITY: ●The nation’s scales are going up…up…up and it’s clear that America has an obesity health crisis. ●Thirteen percent of the adult population were classified as overweight in ●Today, nearly 2 out of 3 adult Americans, 60 percent, are overweight or obese.
3 OBESITY Cont… Health consequences: ●Obesity significantly increases individuals risk for many diseases and health conditions which can be life threatening. ●These individuals also experience a much lower quality of life. ●Obesity has more negative health consequences than smoking, drinking, or poverty and effects more people.
4 OBESITY CONT… Economic consequences: ●The associated health problems of obesity have a tremendous economic impact on the U.S. health care system. ●The economic cost attributable to obesity in the United States is estimated at $100 billion/yr.
5 NULL HYPOTHESIS: Ho: What are the contributing factors to the epidemic of American obesity?
6 THEORY: ●Examine different elements such as behavior, environment, and genetic factors that may have an effect in causing people to be overweight or obese.
7 DATA: ●Developed a survey and recorded 63 observations. Econometrics Analysis Survey: An American Trend that is Reshaping our Nation! 1. What is your age? 2. What is your current weight? 3. How many hours per week do you exercise? 4. How many hours per week are given to leisure activities? (Examples: watching TV, internet surfing, reading) 5. How many times per week do you eat fast foods? 6. Male ___ Female ___ 7. Is over weight hereditary within your family? 8. Do you take any medication that can cause weight gain? (Examples: birth control, depression, anxiety)
8 Data cont… ●Statistics: Mean Standard DeviationMinimumMaximum Weight Age E-hrs/wk L-hrs/wk Food/wk Gender Hereditary Med.& lbs
9 Regression Model: ●Dependent variable is weight. ●Explanatory variables include age, hours of exercise per week, hours of leisure activity per week, number of times eat fast-food per week, gender, obesity hereditary, and take any medications associated with weight gain.
10 First Regression Model: Yi = X 2 – X X X 5 – D D D 8 (Unrestricted equation) R^ Adjusted R^ F Significance F E-14 t Stat P-value Intercept E-06 Age E-hrs/wk L-hrs/wk Food/wk Gender E-06 Hereditary Med& lbs
11 Revised Regression Model: Yi = X2 – X X X5 – D D7 (Restricted equation) R^2 Adjusted R^2 F Significance F Unrestricted Equation E-14 Restricted E-14 Equation
12 Interpreting results: ●Should add variable if: t-test is significant (Ho: coefficient = 0) Adj. R^2 increases t-test was not significant but, Adj. R^2 did increase. ●F test results: P(F7,55> ) <.01 the computed F exceeds the critical F values.
13 Best Regression Equation: Yi = X2 – X X X5 – D D D8 (Unrestricted equation) R^ Adjusted R^ F Significance F E-14
14 Test for Heteroscedasticity using the Park Test: Coefficients t StatP-value Intercept ln predicted ●B2 is not significant, which indicates homoscedasticity
The End. 15 Conclusion : ●My results indicate that a significant relationship exist between the explanatory variables and dependent variable. ●Additional research questions: Are you health conscious? Do you have any conditions that limit physical activity? ●This topic has just begun being researched and there are many obesity externalities that need to be analyzed for better understanding!