Sub: Statistics Topic: Regression

Question:
Barron's conducts an annual review of online brokers, including both brokers that can be
accessed via a Web browser, as well as direct-access brokers that connect customers directly
with the broker's network server. Each broker's offerings and performance are evaluated in six
areas, using a point value of 0-5 in each category.

ClassOf1 provides exert guidance for College, Graduate and High school homework and live online tutoring on subjects like
Finance, Marketing, Statistics, Economics and others. Check out more solved problems in our Solution Library.

The results are weighted to obtain an overall score, and a final star rating, ranging from zero to
five stars, is assigned to each broker. Trade execution, ease of use, and range of offerings are
three of the areas evaluated. A point value of 5 in the trade execution area means the order
entry and execution process flowed easily from one step to the next. A value of 5 in the ease of
use area means that the site was easy to use and can be tailored to show what the user wants
to see. A value of 5 in the range offerings area means that all of the investment transactions
can be executed online. The following data show the point values for trade execution, ease of
use, range of offerings, and the star rating for a sample of 10 of the online brokers that Barron's
evaluated (Barron's, March 10, 2003).

The data set is available in file named Brokers

www.classof1.com
Sub: Statistics Topic: Regression

a. Determine the estimated regression equation that can be used to predict the star rating
given the point values for execution, ease of use, and range of offerings (to 3 decimals).

Rating = ______ + ______ TradeEx + ____ Use + ______ Range

b. Use the t test to determine the significance of each independent variable. What is your
conclusion at the .05 level of significance?
_________________
TradeEx _________________
Use _________________
Range _________________

c. Calculate the coefficient of determination for the equation in part (a) (to 3 decimals).
______
Remove any independent variable that is not significant from the estimated regression
equation. Now, calculate the coefficient of
determination (to 3 decimals).
______
Describe the difference in the fit provided by the two estimated regression equations.

www.classof1.com
Sub: Statistics Topic: Regression

Solution:
Solution:
The information in the given problem can be represented in table as shown below:

Broker Trade Execution Use Range Rating

Wall St. Access 3.7 4.5 4.8 4
E*TRADE (Power) 3.4 3 4.2 3.5
E*TRADE (Standard) 2.5 4 4 3.5
Preferred Trade 4.8 3.7 3.4 3.5
my Track 4 3.5 3.2 3.5
TD Waterhouse 3 3 4.6 3.5
Brown Co 2.7 2.5 3.3 3
Brokerage America 1.7 3.5 3.1 3
Merrill Lynch Direct 2.2 2.7 3 2.5
Strong Funds 1.4 3.6 2.5 2

The regression analyses for the given data is carried out in excel (DataData
AnalysisRegression) and the output is given below:

SUMMARY
OUTPUT

Regression
Statistics
0.941072
Multiple R 549
0.885617
R Square 543
Adjusted R 0.828426
Square 315
Standard Error 0.243099
www.classof1.com
Sub: Statistics Topic: Regression

985
Observations 10

ANOVA
Significanc
df SS MS F eF
2.7454143 0.915138 15.48519 0.0031298
Regression 3 83 128 881 14
0.3545856 0.059097
Residual 6 17 603
Total 9 3.1

Coefficien Standard Upper

ts Error t Stat P-value Lower 95% 95%
-
0.345097 0.5306672 0.650307 0.539580 0.9533989 1.643593
Intercept 005 59 701 378 97 007
0.254822 0.0855553 2.978446 0.024686 0.0454756 0.464168
Trade Execution 076 7 296 197 27 524
-
0.132491 0.1404258 0.943501 0.381848 0.2111177 0.476101
Use 955 6 109 041 46 655
0.458518 0.1231862 3.722159 0.009827 0.1570930 0.759944
Range 96 72 554 255 11 909

From the above coefficient table, the estimated regression equation that can be used to predict
the star rating is given by

Rating = 0.3451 + 0.2548 * Trade execution + 0.1325 * Use + 0.4585 * Range

Coefficient of determination: R-square = 0.8856

www.classof1.com
Sub: Statistics Topic: Regression

The above R-square value indicates 88.56% of the total variation in the dependent
variable is explained by the independent variable.

b.

Coefficients P-value
Intercept (b0 ) 0.34509 0.53958
Trade Execution (b1) 0.25482 0.024686
Use (b2) 0.13249 0.381848
Range (b3) 0.45852 0.009827

From the regression table, we see that the P-value of trade and range are less than 0.05 (level
of significance) hence we conclude that the independent variables trade and range are
significant.
The P-value of the independent variable ‘Use’ is greater than 0.05 which indicates that the
independent variable ‘use’ is not significant.

c)
Since the independent variable (Use) is not significant, it has been eliminated from the
regression equation and the regression analysis is carried out using Excel and we obtained the
following results:

www.classof1.com
Sub: Statistics Topic: Regression

SUMMARY
OUTPUT

Regression
Statistics
0.9320123
Multiple R 96
0.8686471
R Square 07
Adjusted R 0.8311177
Square 09
0.2411857
Standard Error 52
Observations 10

ANOVA
Significanc
df SS MS F eF
2.69280603 1.3464030 23.145777
Regression 2 2 16 81 0.000821
0.40719396 0.0581705
Residual 7 8 67
Total 9 3.1

Coefficien Standard Upper

ts Error t Stat P-value Lower 95% 95%
0.6718441 0.39892179 1.6841500 0.1360252 1.61514
Intercept 63 1 74 49 -0.27146 4
0.2640566 0.08432441 0.0165763 0.46345
Trade Execution 73 1 3.1314381 5 0.064661 2
0.4852712 0.11893470 4.0801485 0.0046888 0.76650
Range 51 2 41 07 0.204035 7

www.classof1.com
 Sub: Statistics Topic: Regression

From the above coefficient table, the estimated regression equation that can be used to predict
the star rating is given by

Rating = 0.6718 + 0.2641 * Trade execution + 0.4853 * Range

Coefficient of determination: R-square = 0.8686

The above R-square value indicates 86.86% of the total variation in the dependent variable is
explained by the independent variable.

Describe the difference in the fit provided by the two estimated regression equations.

By comparing the above output of the regression model with two independent variables with
that of the regression model with three independent variables, we can see that the R-square
value is reduced from 0.8856 to 0.8686. Also going by the ANOVA tables, it is seen that the
model with two independent variables is highly significant as compared to the model with three
independent variables.

** End of the Solution **

ClassOf1 provides exert guidance for College, Graduate and High school homework and live online tutoring on subjects like
Finance, Marketing, Statistics, Economics and others. Check out more solved problems in our Solution Library.

www.classof1.com