Business Analytics

Date: 28-10-
Class & Section: 3 BCom A
24
Reg No Name
2314002 Adarsh .P

-------------------------------------Do not write anything below this -----------------------------------

----------------------------

Marks Secured
Max Marks 20
Initials

Department of Commerce
Business Analytics – COM101-3
Dr. Rahul P
INTRODUCTION:
1. DESCRIPTIVE ANALYTICS :
Descriptive analysis is the process of analyzing
historical data to summarize and understand what has happened in the past. It
involves the use of statistical tools like mean, median, variance, and standard
deviation to describe key features of the data. By presenting data in a structured and
visual manner through charts and graphs, descriptive analysis provides a clear
overview of patterns,trends, and insights, helping to identify areas of improvement or
success.

2. PREDICTIVE ANALYTICS:
Predictive analysis uses historical data to forecast
future outcomes and trends. By applying techniques such as regression analysis
it identifies patterns and relationships between variables that allow predictions to
be made about future behavior. This type of analysis is valuable in helping
businesses and individuals make informed decisions, as it anticipates potential
results based on past data, such as predicting sales based on marketing spend or
forecasting exam scores based on
study hours.

3. PRESCRIPTIVE ANALYTICS:

Predictive analysis uses historical data to forecast

future outcomes and trends. By applying techniques such as regression analysis, it
identifies patterns and relationships between variables that allow predictions to be
made about future behavior. This type of analysis is valuable in helping businesses
and individuals make informed decisions, as it anticipates potential results based on
past data, such as predicting sales based on marketing spend or forecasting exam
scores based on study hours.
DESCRIPTIVE ANALYSIS:

Q. Analyze the relationship between observation and no. of units sold (sales) by applying
descriptive analysis techniques. Calculate and interpret the measures of central tendency
(mean, median, mode), measures of dispersion (range, variance, standard deviation), and
shape of the distribution (skewness, kurtosis) for the average call duration. Additionally,
visualize the data using relevant charts such as histograms, box plots, or scatter plots to
uncover trends and patterns. What insights can you derive from this analysis about how
changes with units sold ?

Units
observation
Sold
1 52
2 47
3 55
4 53
5 58
6 52
7 50
8 57
9 60
10 54
11 49
12 53
13 55
14 50
15 57
16 61
17 56
18 53
19 48
20 54
21 59
22 52
23 49
24 58
25 60
26 55
27 51
28 54
29 53
30 57
 1. Measures of central Tendency:

mean 54.0667
median 0.68133
mode 53

a. Mean (54.0667): The mean, or average, is relatively high at 54.07. This suggests that
the dataset contains large values, which may be inflating the mean. When the mean is
much higher than the median, it often indicates the presence of extreme values or
outliers that skew the data to the right.
b. Median (0.68133): The median is very low at approximately 0.68, much lower than
the mean. The median represents the middle value of the data when sorted in
ascending order, so this low value suggests that a majority of the data points are small
numbers. The significant difference between the mean and median suggests a
positively skewed distribution, where the bulk of the data lies near the lower end, but
there are some very large values pulling the mean upward.
c. Mode (53): The mode is 53, which means that 53 is the most frequently occurring
value in the dataset. This is close to the mean but still much higher than the median,
reinforcing the idea of a skewed distribution.

2. Measures of dispersion:

Range 14
variance 13.92644
Standard
Deviation 3.731814

a. Range (14): The range is the difference between the maximum and minimum values in the
dataset, which is 14. This value indicates a relatively small spread from the lowest to the
highest value, especially given the high mean (54.07) noted previously. This suggests that
although there are some larger values, most of the data is concentrated within a narrow.

b. Variance (13.93): Variance measures how much the data points deviate from the mean on
average. A variance of 13.93 indicates moderate variability. In the context of the mean, this
level of variance is not extremely high, suggesting that while there is some spread around the
mean, the data points aren’t highly dispersed.
c. Standard Deviation (3.73): The standard deviation, which is the square root of the
variance, provides a more interpretable measure of spread, indicating that data points are
typically around 3.73 units away from the mean on average. Given the earlier statistics, this
moderate standard deviation suggests that despite the skew, most of the data points fall within
a limited range around the mean.

3. Measures of shape of distribution:

skewness 0.033567
kurtosis -0.7213

a. Skewness (0.0336): Skewness measures the asymmetry of the distribution. A

skewness close to zero (in this case, 0.0336) suggests that the data is
approximately symmetric, with no significant skew to the left or right. This is
interesting because the earlier statistics (mean much higher than median)
suggested some asymmetry. The low skewness value may indicate that while the
distribution has a few large values, these aren’t extreme enough to create a
pronounced skew.

Histogram
8

6
Frequency

4
Series1
2

0
45 47 49 51 53 55 57 59 61 63
Bins

b. Kurtosis indicates the "tailedness" of the distribution, with positive values

indicating heavy tails (more extreme outliers) and negative values indicating
lighter tails. A kurtosis of -0.7213 suggests a distribution with lighter tails than a
normal distribution, often referred to as platykurtic.
PREDICTIVE ANALYSICS:

Q. Using the dataset provided, which contains the yearly revenue and gross profit of a
company from 1995 to 2024, conduct a regression analysis in MS Excel to explore the
relationship between Sales and Advertisement Expense.

Units
Advertisement
Observation Sold
expenditure (x)
(y)

1 15 52
2 18 55
3 20 57
4 22 60
5 24 63
6 19 56
7 21 59
8 17 53
9 23 62
10 16 54
11 18 56
12 25 65
13 28 69
14 30 72
15 27 68
16 26 66
17 29 70
18 22 61
19 19 55
20 21 58
21 24 63
22 25 64
23 30 71
24 27 67
25 16 53
26 20 57
27 23 61
28 28 68
29 17 54
30 15 51
 1. REGRESSION STATISTICS:

The Multiple R value of 0.99 indicates a strong connection, implying that as revenue
improves, so does gross profit. The R Square value of 0.99 indicates that changes in sales
account for approximately 85%of the variation in advertisement account, indicating a strong
model fit. The Standard Error of 6.086 reflects the average distance between the observed
values and the regression line, and it serves as a measure of the model's prediction accuracy.
Overall, these results suggest that sales is a significant predictor of advertisement , with a
strong explanatory power.

Regression Statistics
Multiple R 0.995223
R Square 0.99047
Adjusted R
Square 0.954755
Standard Error 6.086513
Observations 29

80

70

60

50

40

30

20

10

0
0 5 10 15 20 25 30 35
a. Multiple R (0.9952): This is the correlation coefficient between the observed and
predicted values. A Multiple R of 0.9952 indicates a very strong positive linear
relationship between the independent and dependent variables in the model. A value
this close to 1 suggests that the model's predictions are highly correlated with the
actual data.
b. R Square (0.9905): R Square, or the coefficient of determination, measures the
proportion of variance in the dependent variable that can be explained by the
independent variable(s) in the model. An R Square of 0.9905 indicates that
approximately 99.05% of the variability in the dependent variable is explained by the
model. This is a very high value, suggesting that the model fits the data exceptionally
well.
c. Adjusted R Square (0.9548): Adjusted R Square is a modified version of R Square
that adjusts for the number of predictors in the model, penalizing for adding variables
that don’t improve the model significantly. An Adjusted R Square of 0.9548 (95.48%)
still represents a strong model, though it is slightly lower than R Square. This
decrease indicates that while the model is still very effective, there may be a slight
overfitting or redundancy in the predictors, though it’s not very significant

d. Standard Error (6.0865): The standard error of the regression measures the average
distance that the observed values fall from the regression line. A smaller standard
error indicates that the data points are closer to the fitted line, suggesting a more
accurate model. A standard error of 6.0865 indicates a reasonably precise model,
though the acceptable range would depend on the context and scale of the dependent
variable.
 2. ANOVA ( Analysis of variance )

df SS MS F Significance F
Regression 1 107802.7 107802.7 2909.998 4.907E-29
Residual 28 1037.278 37.04563
Total 29 108840

The Regression Sum of Squares (SS) is 107802.7 which is significantly bigger than the
Residual SS of 1037.278 indicating that changes in Revenue account for the majority of the
variability in Gross Profit. The F-value of 2909.998is high, and the exceptionally low
Significance F 4.907E-29 value demonstrates that the link between sales and advertisenment
expense is statistically significant. Furthermore, the R-squared value of 0.75 validates the
claim that sales accounts for 75% of the variance in advertisement demonstrating a strong
and meaningful linear association between the two variables.

3. COEFFICIENT TABLE :

P-
Coefficien Standard valu Lower Upper Lower Upper
ts Error t Stat e 95% 95% 95.0% 95.0%
Interce #N/
pt 0 #N/A #N/A A #N/A #N/A #N/A #N/A
2.668582 0.04946 53.9443 7.65 2.567249 2.76991 2.56724 2.769915
15 375 913 97 E-30 457 529 946 294

a. Coefficients: The coefficient for the intercept term is 15, indicating the expected value of
the dependent variable when all predictors are zero.

b.Standard Error: The standard error for the intercept term is 2.668582375, giving an idea
of the precision of the coefficient estimate. Lower standard errors indicate more precise
estimates.

c. t Stat: The t-statistic for the intercept is 53.944397, suggesting that the coefficient is
highly significant. A high t-statistic generally indicates a coefficient significantly different
from zero.
d. P-value: The p-value is very small (7.65E-30), indicating that the intercept term is
statistically significant. A small p-value (typically less than 0.05) suggests that we can reject
the null hypothesis (which would state that the intercept is zero).

e. Confidence Intervals:

 Lower 95% and Upper 95% bounds provide a confidence interval for the intercept.
This interval is 2.567249457 to 2.76991529, indicating that we are 95% confident that
the true value of the intercept lies within this range.

PRESCRIPTIVE ANALYTICS:

Q. Tata Motors manufactures three types of cars: X1, X2, and X3. Each model requires
specific amounts of time for assembly, painting, and inspection. Tata Motors has a limited
amount of resources in each area, and the objective is to maximize the weekly profit.
Producing one unit of model X1 yields a profit of ₹5,000, one unit of model X2 yields
₹6,000, and one unit of model X3 yields ₹8,000. The company has 4 assembly lines, each
available for 50 hours a week, resulting in a total of 200 hours. Each unit of X1, X2, and X3
requires 5, 7, and 8 hours of assembly time, respectively. In the paint department, there are 3
paint booths, each operational for 40 hours a week (120 hours total). Model X1 requires 3
hours of paint time per unit, X2 requires 4 hours, and X3 requires 2 hours. Lastly, there are 2
inspection lines, each running 30 hours per week, for a total of 60 hours. Each unit of X1
requires 1 hour of inspection, X2 requires 2 hours, and X3 requires 3 hours. The objective is
to determine the optimal number of each car model to produce weekly (X1, X2, and X3) to
maximize profit, while staying within the constraints of assembly, painting, and inspection
time

Solution:

objective function:

maximize = 5000x1+6000x2+8000x3

Assembly time constraint :

5x1+7x2+8x3 ≤200

Paint time constraint:

 3x1+4x2+2x3≤120

Inspection time constraint :

x1+2x2+3x3≤60

4. Non-negativity Constraints :

x1≥0, x2≥0 , x3≥0

VARIABLES
X1 17.14286
X2 0
X3 14.28571

OBJECTIVE FUNCTION

maximize z 200000

CONSTRAINTS

200 <= 200

80 <= 120
60 <= 60

17.14285714 >= 0
0 >= 0
14.28571429 >= 0

∴ To get maximum profit of $ 200000 while following all the constraints related
to raw material, labor and machine hours availability, the company should
produce 6 units of product A, 8 units if product B and 8 units of product C.