Professional Documents
Culture Documents
Work 2 Statistics For Economists
Work 2 Statistics For Economists
TABLE N°1
Descriptive statistics: Florida, New York, North Carolina, for
people in good health.
Interpretation
The geographical location of New York has a higher level of depression
than the other locations.
TABLE N°2
Descriptive statistics: Florida, New York, North Carolina, for
people suffering from chronic conditions such as arthritis,
hypertension or heart disease.
Interpretation
The geographical location of New York has a higher level of depression
than the other locations.
2. Use analysis of variance for both data sets. In each case give the
hypotheses to be tested. What are your conclusions?
Hypothesis
H0: U1=U2=U3 (all means are equal)
H1: U1#U2#U3 (not all averages are equal)
Significance level: α = 0.05
TABLE N°3
Analysis of Variance
Sourc G SC MC F- p-
e L Adjust. Adjust. value value
Factor 2 61.03 30.517 5.24 0.008
Error 57 331.90 5.823
Total 59 392.93
Conclusions: The null hypothesis is rejected then not all means are
equal. at least two of the population means have a different value.
TABLE N°4
Analysis of Variance
Sourc G SC MC F- p-
e L Adjust. Adjust. value value
Factor 2 17.03 8.517 0.71 0.494
Error 57 679.70 11.925
Total 59 696.73
Conclusions: The null hypothesis is accepted then all means are equal.
3. If necessary, use inferences about the means of each of the treatments.
What are the conclusions?
Conclusion:
In the first case the null hypothesis is rejected, in the second case the
null hypothesis is accepted.
In the geographical location of New York a higher level of depression is
obtained, in both cases, since they have a higher mean.
Hypothesis
H0: errors have normal distribution
H1: errors do not have normal distribution
Level of significance:
α = 0.05
Test statistic:
Anderson-Darling normality test
AD= 0.348
P-Value: 0.472
Decision: the null hypothesis is not rejected.
Conclusion: with a significance of 5% it is concluded that the errors have a
normal distribution (the assumption is fulfilled).
HOMOGENEITY OF VARIANCES
Hypothesis:
H0: variances are equal (homogeneous)
H1: at least one variance is different
Level of significance:
α = 0.05
Test statistic:
Barlett: 0.42
P value: 0.995
Decision: the null hypothesis is not rejected.
Conclusion: with a significance of 5% it is concluded that the variances are
homogeneous (the assumption is met).
Administrative report
1. Use descriptive statistics to summarize the data.
Hypothesis:
Ho: no effect of the type of salesperson on the average salary of
employed individuals.
H1: there is an effect of the type of salesperson on the average wage of
employed individuals.
Level of significance:
α = 0.05
TABLE N°5
Analysis of Variance
F- p-
Source GL SC Adjust. MC Adjust. value value
Position 1 9515793950 9515793950 751.36 0.000
Experience 2 1668100099 834050050 65.86 0.000
2 1352066184 676033092 53.38 0.000
Position*Experience
Error 114 1443782758 12664761
Total 119 1397974299
2
Test statistic:
F= 751.36
P-Value = 0.000
Decision: the null hypothesis is rejected.
Conclusion: with a significance of 5% it is concluded that there is an
effect of some type of salesperson (internal, external) on the average
salary of employed individuals.
TABLE N°5
Group information using the
Tukey's method and 95% confidence
Positio
n N Media Grouping
Outside 60 73830.4 A
Inside 60 56020.5 B
Means that do not share a letter are significantly different.
Hypothesis
H0: U1=U2 ( means are equal)
H1: U1#U2 (averages are not equal)
Significance level: α = 0.05
Conclusion: the null hypothesis is rejected, the means are not equal.
Hypothesis:
Ho: there is no effect of years of experience on the average wage of
employed individuals.
H1: There is an effect of years of experience on the average wage of
employed individuals.
Level of significance:
α = 0.05
TABLE N°6
Analysis of Variance
F- p-
Source GL SC Adjust. MC Adjust. value value
Position 1 9515793950 9515793950 751.36 0.000
Experience 2 1668100099 834050050 65.86 0.000
2 1352066184 676033092 53.38 0.000
Position*Experience
Error 114 1443782758 12664761
Total 119 1397974299
2
Test statistic:
F= 65.86
P-Value = 0.000
Decision: the null hypothesis is rejected.
Conclusion: With a significance of 5%, it is concluded that there is an
effect of years of experience on the average salary of employed
individuals.
Hypothesis
H0: U1=U2=U3 (all means are equal)
H1: U1#U2#U3 (not all averages are the same)
Significance level: α = 0.05
Conclusion: the null hypothesis is rejected, the means are not equal.
7. With 0.05 as the significance level, test for significant differences due to
type of salesperson, years of experience or interaction.
Hypothesis:
Ho: there is no interaction between type of salesperson and years of
experience influencing the average salary of employed individuals.
H1: There is interaction between the type of salesperson and years of
experience that influence the average salary of employed individuals.
Level of significance:
α = 0.05
TABLE N°5
Analysis of Variance
F- p-
Source GL SC Adjust. MC Adjust. value value
Position 1 9515793950 9515793950 751.36 0.000
Experience 2 1668100099 834050050 65.86 0.000
2 1352066184 676033092 53.38 0.000
Position*Experience
Error 114 1443782758 12664761
Total 119 1397974299
2
Test statistic:
F= 53.38
P-Value = 0.000
Decision: the null hypothesis is rejected.
Conclusion: With a significance of 5%, it is concluded that there is an
interaction between the type of salesperson and the years of experience
that influence the average salary of the employed individuals.