DS 303 STUDY GUIDE MIDTERM

*Regression analysis was applied between sales (in $1000s) and advertising (in $100s), and the following regression function was obtained. ^y= 500 + 4x Based on the above estimated regression line, if advertising is $10,000, what is the point estimate for sales (in dollars)?

$900,000

*We ran a regression to find out how number of crimes (y) is a affected by unemployment (x1), number of police officers (x2) and population (x3). We collected data for 64 metropolitan areas and found the following results. What is 99% confidence interval for unemployment rate, B3?

(0.021, 0.062)

*US federal government was evaluating how total trac fatalities (y) are impacted by: vehicle miles travelled in billions (x1), unemployment rate in percent (x2) Percentage of population aged between 14 to 24 and (x3) state population in hundred thousand (x4) They use 1200 observation to run a regression with the following results: What is 95% confidence interval for unemployment rate, B2?

(5.982, 25.409)

*In a multiple regression model, the disturbance term is assumed to, 1. have a mean of 1. 2. have a variance of zero. 3. have no distribution. 4. be normally distributed

Assumed to be normally distributed

A regression analysis between demand (y in 1000 units) and price (x in dollars) resulted in the following equation: ŷ = 9 − 3x The above equation implies that if the price is increased by $1, the demand is expected to _____. A) increase by 6 units B) decrease by 3 units C) decrease by 6,000 units D) decrease by 3,000 units

D) decrease by 3,000 units

Which of the following is true about Residuals? A) Lower is better B) Higher is better C) A or B depend on the situation D) None of the above

A) Lower is better

Scatter plots in regression will not tell us exact correlation but will allow us to infer the direction and probable strength of relationship. A) True] B) False

A) True

If two variables, x and y, have a strong correlation, then _____. A) there may or may not be any causal relationship between x and y B) x causes y to happen C) y causes x to happen D) None of the answers is correct

A) there may or may not be any causal relationship between x and y

If we run a regression to study the effects of study hours (x) on GPA (y). Then GPA is_______________ variable, and study hours is ______________variable. A) Explanatory; Response B) Independent; Dependent C) Response; Explanatory D) Independent; Explanatory

C) Response; Explanatory

In regression analysis, if the dependent variable is measured in dollars, the independent variable _____. A) must also be in dollars B) must be in some unit of currency C) can be any units D) cannot be in dollars

C) can be any units

A regression model between sales (y in $1000), unit price (x1 in 100 dollars), and television advertisement (x2 in 1000 dollars) resulted in the following function: ŷ = 7 - 3x1 + 5x2 The coefficient of the advertisement indicates that holding unit price constant, if the expenditure is: A) increased by $1, sales are expected to decrease by $5. B) increased by $1000, sales are expected to decrease by $5. C) increased by $1, sales are expected to increase by $5. D) increased by $1, sales are expected to increase by $5000.

C) increased by $1, sales are expected to increase by $5

A multiple regression model has A) only one independent variable B) more than one dependent variable. C) more than one independent. D) at least two dependent variables

C) more than one independent

Fitted error or residual is the difference between A) observed values of the independent variable and the predicted values of the independent variable B) actual values of the independent variable and the predicted values of the dependent variable C) observed values of the dependent variable and the predicted values of the dependent variable D) None of the answers is correct.

C) observed values of the dependent variable and the predicted values of the dependent variable

*Equation (^y)= b0+b1X1 depicts 1. Difference between observed and predicted values 2. Observed relationship in the sample 3. True relationship in the population 4. None of the above

Observed relationship in the sample

*Ameren utility wanted to nd the e ects of temperature (x) on the natural gas consumption (y). Using 16 observations from Macomb neighborhood it ran a regression and reported following results. Identify the standard error of regression?

Reported in `Standard Error' in Regression Statistics - 0.339

*A regression model between tickets sold (y in 1000), the temperature in Fahrenheit (x1) and the total snowfall in inches (x2)resulted in the following function: y= 5688 +235x1 + 38x2 The coefficient of the temperature indicates that if the temperature: 1. increases by 1 Fahrenheit, holding snowfall constant the ticket sales are expected to increase by $235. 2. decreases by 1 Fahrenheit, holding snowfall constant the ticket sales are expected to increase by $235000. 3. increases by 1 Fahrenheit, holding snowfall constant the ticket sales are expected to increase by $235000. 4. increases by 1 Fahrenheit, holding snowfall constant the ticket sales are expected to increase by $235

increases by 1 Fahrenheit, holding snowfall constant the ticket sales are expected to increase by $235000.

The least squares criterion is _____.

minE(yi-ybar)^2

*se=sq rt [(sigma(yi-y^)^2/(n-K-1)=sq rt [(sigma(e^i)^2/(n-K-1) defines 1. standard error of regression in multiple regression only 2. standard error of regression in multiple and simple regression 3. square of standard error of regression in multiple regression only 4. square of standard error of regression in simple and multiple regression

standard error of regression in multiple and simple regression