Regression Analysis of Television Channels

...

Table 1: Variables of the dataset

-

Descriptive Statistics

The dataset was analysed using STATA (Version 14.1). To begin with, the descriptive statistics of the dataset were found in order to summarize the dataset. Specifically, the mean and the standard deviation, as well as the minimum and the maximum values were found.

As mentioned before, the structure of the presentation comprises of the STATA command mentioned first, followed by the result generated by the software.

STATA command:

- . summarize & describe

STATA result:

Variable

Obs

Mean

Std. Dev.

Min

Max

saleprice

31

1981.774

2530.735

13

12634

marketretailsales

31

1873.516

2037.011

85

8472

marketbuyingincome

31

2473.903

2729.924

136

13733

tvhomesstation

31

206.129

288.6203

15

1553

networkhourlyrate

31

684.0323

670.0242

60

3500

nationalspotrate

31

143.7742

139.2201

25

750

ageofstation

31

0.7741935

.4250237

0

1

numberoftieswithmajornetworks

31

0.9032258

.8308565

0

2

percentofmarketpopulationinurban

31

52.97419

17.52246

23.1

88.3

Table 2: Descriptive statistics of the dataset

By observing the difference between the mean value and the range of each variable, one has to point out that there exist a few outliers in the dataset, perhaps a larger TV station than the others (see graphs below).

-

Null Hypothesis Testing

One major difference between the TV Stations described in the dataset, is their age. Some of the included TV Stations were established before January 1st, 1952 and have an ageofstation variable value of 0 (7 in total). On the other hand, TV stations with a first-aired date after January 1st 1952 (24 in total) have a value of 1.

An interesting aspect is to examine whether the mean of the sale prices of the TV stations with first-air date before 1952 differs greatly from the mean of the sale prices of TV station first aired after 1952. In order to inspect the existence of such a difference, a two-sample t-test is used, with the assumption that the two groups (first-air date before and after 1952) have equal variances.

STATA command:

- . ttest saleprice, by(ageofstation)

STATA result:

Two-sample t test with equal variances

Group

Obs

Mean

Std. Err.

Std. Dev.

[95% Conf. Interval]

0

7

3777.714

1590.748

4208.723

-114.705

7670.134

1

24

1457.958

320.4827

1570.038

794.9893

2120.927

combined

31

1981.774

454.5334

2530.735

1053.493

2910.055

diff

2319.756

1018.33