Business Startup Costs and Sales Analysis

This assignment delves into the analysis of business startup costs using statistical methods. Students construct frequency tables, histograms, and conduct ANOVA to examine differences in startup costs across businesses. Additionally, they perform regression analysis to understand the relationship between startup costs and annual sales, interpreting coefficients and assessing model fit. The assignment highlights key observations about distribution patterns and the significance of independent variables in predicting annual sales.

STATISTICS
Student Name & ID
[Pick the date]
Task 1
1

Given data set
1. Requisite statistical parameters are computed below.
2. Construction frequency and relative frequency table and histogram
2

(a) Frequency and relative frequency table by taking class as 0 to 30
3

(b) Histogram for relative frequency of the variables is shown below:
4

5

3. Key observations are summarized below.
 The distribution in relation to the startup costs is non-normal for all businesses. The
evidence in this regards is derived from skew not being zero.
 Further, the setup costs for most businesses tend to be larger than $ 30,000. A noticeable
exception in this regard is pet business.
6

4. Hypothesis testing
H0 : No significnt difference present ∈starting costs for the business types .
H1 : Minimum one of the bus iness has significant difference∈starting costs form other businesses .
ANOVA Single Factor
It is observed that the p value magnitude is lesser than α and hence it leads to null hypothesis
rejection. As a conclusion, it would be fair that business starting costs tend to show significant
difference.
TASK 2
1) Based on the given data given, the excel output through regression is outlined below.
7

Regression equation:
(1) Measure of fit: R2
Value: 0.9932
Implication: 99.32% of the changes in annual sales are accounted for by the predictor
variables considered.
Conclusion: Good Fit
(2) Hypothesis Formulation
Regression Output
8

The corresponding p value from above = 0.00. This implies that null hypothesis would be
rejected thus indicating significance of atleast one amongst the five independent variables.
(3) Slope coefficient interpretation
(4) Confidence level from regression output is tabled below.
(5) Formulation of Hypothesis
The table shown below highlights the relevant output in relation of hypothesis testing.
9

(6) The model has no changes to be made considering that all variables are significant.
(7) The regression equation originally derived is indicated below.
The given inputs are fed into the above regression equation to yield the following result.
.
10