Formulate a testable hypothesis (e.g. higher temperature will increase sales of
ID: 3179533 • Letter: F
Question
Formulate a testable hypothesis (e.g. higher temperature will increase sales of Checker's Pizza) Why did your group choose this variable? What type of relationship does this variable has with DV? Did the model improve, deteriorated, or stayed the same? How do you know? What insight does the t-stat provides? Credibility to the P Is the new IV statistically significant at 1%? 2%? or 5%? Would be 36% higher not statistically significant Did the coefficient, t-ratio, p-value, R2 changed for other IVs? What might be the justification for the change They all changed 2.552 2 % 24.6- 18 degree 3 are under and three are over What does the new IV tell the decision makers? The change in price for the competitors will impact the sales for checkers Any other important information that provide insights? How do you plan to apply the concepts and lesson learned from this consulting project? Final Presentation + 1 pager response for your classmatesExplanation / Answer
(a) In the given multiple regression model, temperature variable could be added. In order to check whether its cofficient will have significance impact on the explanatory variable ( sales ). We will have the hypothesis as
H0 : cofficient of temperature variable is equal to zero
Ha : cofficient of temperature variable is diffferent from zero
(b) This variable can be chosen to improve the accuracy of the model. This can be measured with cofficient of determination (R2) , whih helps to explain the proportion of variation in the response variable explained by the explanatory variable. If addition of a new variable increases the R2 then it is benificial to have it.
(c) This variable may have positive or negative relationship with the dependent variable depending the tendency of consumers in different temperatures . Some can have it difficult to cook in low temperatures and go for pizza (in this case cofficient will be negative) , some can have the opposite and reverse will be the sign of cofficient of temperature variable.
(d)This can be measured with cofficient of determination (R2) , whih helps to explain the proportion of variation in the response variable explained by the explanatory variable. If addition of a new variable increases the R2 then model is improved.
(e) t-stat helps us test whether cofiicient is significant or not when compared to critical value from the t- tables.