Author Question: If time-series data exhibit a seasonal pattern, which of the following approaches could be used to ... (Read 118 times)

stevenposner

  • Hero Member
  • *****
  • Posts: 608
If time-series data exhibit a seasonal pattern, which of the following approaches could be used to compute season indexes?
 
  A) The exponential smoothing technique
  B) The multiplicative model
  C) The Pearson product-moment approach
  D) None of the above

Question 2

What factors are of importance to an analyst when linear regression analysis is used for descriptive purposes?
 
  What will be an ideal response?



lorealeza

  • Sr. Member
  • ****
  • Posts: 345
Answer to Question 1

B

Answer to Question 2

The primary objective when using regression analysis for descriptive purposes is to measure the relationship between the dependent and independent variables. There are several statistics that are useful for this purpose. The first is the correlation coefficient. The closer this measure is to positive 1.0 or minus 1.0, the stronger is the linear relationship. The R-square value measures the percentage of variation in the dependent variable that is explained by the independent variable. Values close to 1.0 (or 100 percent) indicate that a strong linear relationship exists.

However, the primary statistics of interest are the regression coefficients. The intercept value (b0 ) can be interpreted to be the average value of y when x = 0.0. Note: This interpretation is valid only when the data used to develop the regression equation can legitimately have values of the x variable equal to zero. If not, then the intercept has no particular interpretation. The regression slope coefficient (b1 ) is usually of prime importance in a descriptive analysis. We interpret it to mean the average change in y for a one-unit change in x. We are particularly interested in the sign on the slope coefficient and the actual value of the coefficient too. We remember that the value of b1 is a point estimate and, thus, subject to sampling error. Therefore, we would likely develop a confidence interval estimate for the true population regression slope coefficient. If this interval does not cross zero, we say that there is a significant linear relationship.



Related Topics

Need homework help now?

Ask unlimited questions for free

Ask a Question
 

Did you know?

Medication errors are three times higher among children and infants than with adults.

Did you know?

The human body's pharmacokinetics are quite varied. Our hair holds onto drugs longer than our urine, blood, or saliva. For example, alcohol can be detected in the hair for up to 90 days after it was consumed. The same is true for marijuana, cocaine, ecstasy, heroin, methamphetamine, and nicotine.

Did you know?

Fungal nail infections account for up to 30% of all skin infections. They affect 5% of the general population—mostly people over the age of 70.

Did you know?

The average adult has about 21 square feet of skin.

Did you know?

People about to have surgery must tell their health care providers about all supplements they take.

For a complete list of videos, visit our video library