Scribd
Upload
68 views1 page

Regression

The document contains three sections describing analyses of time series data: A) A multiple regression model was fitted to study the effect of advertising (X1) and market share (X2) on product sales (Y). The model estimates were β1=3.408 and β2=1.175. B) An exponential trend model Yt=abt was fitted to sales data from 1961-1981. C) A linear trend model was fitted to annual sales data from 1980-1990 for a product introduced 10 years ago. Forecasts were made for 1990 and 1992 and the error estimated for 1990.

Uploaded by

kabo busanang
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
68 views1 page

Regression

The document contains three sections describing analyses of time series data: A) A multiple regression model was fitted to study the effect of advertising (X1) and market share (X2) on product sales (Y). The model estimates were β1=3.408 and β2=1.175. B) An exponential trend model Yt=abt was fitted to sales data from 1961-1981. C) A linear trend model was fitted to annual sales data from 1980-1990 for a product introduced 10 years ago. Forecasts were made for 1990 and 1992 and the error estimated for 1990.

Uploaded by

kabo busanang
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
You are on page 1/ 1

A) In a study of sales of a certain product (Y) by market area, information on advertising expenditure and

market share were collected. To find out the effect that advertising (X 1) and market share (X 2) has on
amount of sales, a multiple regression model was fitted using data from 10 companies, summarized
below.

x1y=21491.33 x2y=42466.38 x1=136.2 x2 =271.75 y=945

i) If the estimated values of β^ 1=3 . 408 and β^ 2=1 . 175 , find the β^ 0
^ β^ 2
ii) Interpret each parameter estimate β 1 and
iii) Find the estimated value of sales when x1 = 24 and x2 = 45

B) The following summary statistics were obtained from a time series data that ran from 1961 to 1981
20 20 20

∑ t=0 ∑ y =5490. 4
t=0
t ∑ log y t =45 .373
t =0
∑ t log y t =67 .160
t=0

∑ t 2=770
Fit an exponential trend model of the type: Y t = abt

[8+5+5+12]

The sale and operating costs of a company in Safire Tyre Services in (in thousands pula) for the
years 1998-2007 were 120, 239, 201, 152, 358, 250, 333, 321, 335, and 387.
B2.1 Plot the time series data and comment on the plot.
B2.2 Would it be more appropriate to use the linear or polynomial trend? Explain.
B2.3 Fit a linear trend equation if summary of data is given as
2007 2007
∑ t=0 ∑ Y t =2696 ∑ Y t t=4110 ∑ t 2=330
t=1998 t=1998
B2.4 Interpret each linear regression parameter estimate.
B2.5 Estimate the cost in 2009 and 2010.
B2.6 What would you recommend to the manager of this company?

A marketing researcher studied annual sales of a product that had been introduced 10 years ago.
The data were as follows.

Years: 1980 1981 1982 1983 1984 1985 1986 1987 1989 1990
Sales: 99 136 165 175 220 235 285 295 370 385

(i) Draw a time series plot for the above data, and give reasons why a simple linear trend model
would best describe the data.
(ii) Fit a simple linear trend model of the data and give interpretations of the coefficients.
Assume that 
t0
(iii) Give the forecasted values for the years 1990, 1992, and give estimates for the error in 1990.

You might also like