UNIVERSITY TECHNOLOGY MARA CAMPUS BANDARAYA MELAKA

FACULTY OF BUSINESS MANAGEMENT

BA240 - BACHELOR OF BUSINESS ADMINISTRATION (HONS.)

MARKETING

MGT555

INTRODUCTION TO BUSINESS ANALYTICS

GROUP ASSIGNMENT

GROUP:

M1BA240 3C

PREPARED BY:

NO STUDENT’S NAME MATRIC NUMBER

1 NOR SHAFIKAH BINTI NOR AZMAN 2022698086

2 NUR AFRINNA HANANI BINTI ROZAIDI 2022698192

3 FIRAS DANIAL BIN FAHMI 2022697894

4 MOHAMMAD NURAFIQ NABIL BIN MOHD 2022455732

NOOR

PREPARED FOR

DR UMMI KALSUM BINTI HASSIAN

SUBMISSION DATE

18 JANUARY 2024
QUESTION
Obtain any dataset from a business organization (e.x. marketing related data, financial data,
human resources data, production data, customer survey data etc) OR any dataset from the
data source listed below.
Price
Transaction Customer Product per Total
ID Date ID Gender Age Category Quantity Unit Amount
1 24/11/2023 CUST001 Male 34 Beauty 3 50 150
2 27/2/2023 CUST002 Female 26 Clothing 2 500 1000
3 13/1/2023 CUST003 Male 50 Electronics 1 30 30
4 21/5/2023 CUST004 Male 37 Clothing 1 500 500
5 6/5/2023 CUST005 Male 30 Beauty 2 50 100
6 25/4/2023 CUST006 Female 45 Beauty 1 30 30
7 13/3/2023 CUST007 Male 46 Clothing 2 25 50
8 22/2/2023 CUST008 Male 30 Electronics 4 25 100
9 13/12/2023 CUST009 Male 63 Electronics 2 300 600
10 7/10/2023 CUST010 Female 52 Clothing 4 50 200
11 14/2/2023 CUST011 Male 23 Clothing 2 50 100
12 30/10/2023 CUST012 Male 35 Beauty 3 25 75
13 5/8/2023 CUST013 Male 22 Electronics 3 500 1500
14 17/1/2023 CUST014 Male 64 Clothing 4 30 120
15 16/1/2023 CUST015 Female 42 Electronics 4 500 2000
16 17/2/2023 CUST016 Male 19 Clothing 3 500 1500
17 22/4/2023 CUST017 Female 27 Clothing 4 25 100
18 30/4/2023 CUST018 Female 47 Electronics 2 25 50
19 16/9/2023 CUST019 Female 62 Clothing 2 25 50
20 5/11/2023 CUST020 Male 22 Clothing 3 300 900
21 14/1/2023 CUST021 Female 50 Beauty 1 500 500
22 15/10/2023 CUST022 Male 18 Clothing 2 50 100
23 12/4/2023 CUST023 Female 35 Clothing 4 30 120
24 29/11/2023 CUST024 Female 49 Clothing 1 300 300
25 26/12/2023 CUST025 Female 64 Beauty 1 50 50
26 7/10/2023 CUST026 Female 28 Electronics 2 500 1000
27 3/8/2023 CUST027 Female 38 Beauty 2 25 50
28 23/4/2023 CUST028 Female 43 Beauty 1 500 500
29 18/8/2023 CUST029 Female 42 Electronics 1 30 30
30 29/10/2023 CUST030 Female 39 Beauty 3 300 900
31 23/5/2023 CUST031 Male 44 Electronics 4 300 1200
32 4/1/2023 CUST032 Male 30 Beauty 3 30 90
33 23/3/2023 CUST033 Female 50 Electronics 2 50 100
34 24/12/2023 CUST034 Female 51 Clothing 3 50 150
35 5/8/2023 CUST035 Female 58 Beauty 3 300 900
36 24/6/2023 CUST036 Male 52 Beauty 3 300 900
37 23/5/2023 CUST037 Female 18 Beauty 3 25 75
38 21/3/2023 CUST038 Male 38 Beauty 4 50 200
39 21/4/2023 CUST039 Male 23 Clothing 4 30 120
40 22/6/2023 CUST040 Male 45 Beauty 1 50 50
41 22/2/2023 CUST041 Male 34 Clothing 2 25 50
42 17/2/2023 CUST042 Male 22 Clothing 3 300 900
43 14/7/2023 CUST043 Female 48 Clothing 1 300 300
44 19/2/2023 CUST044 Female 22 Clothing 1 25 25
45 3/7/2023 CUST045 Female 55 Electronics 1 30 30
46 26/6/2023 CUST046 Female 20 Electronics 4 300 1200
47 6/11/2023 CUST047 Female 40 Beauty 3 500 1500
48 16/5/2023 CUST048 Male 54 Electronics 3 300 900
49 23/1/2023 CUST049 Female 54 Electronics 2 500 1000
50 24/8/2023 CUST050 Female 27 Beauty 3 25 75
51 2/10/2023 CUST051 Male 27 Beauty 3 25 75
52 5/3/2023 CUST052 Female 36 Beauty 1 300 300
53 13/7/2023 CUST053 Male 34 Electronics 2 50 100
54 10/2/2023 CUST054 Female 38 Electronics 3 500 1500
55 10/10/2023 CUST055 Male 31 Beauty 4 30 120
56 31/5/2023 CUST056 Female 26 Clothing 3 300 900
57 18/11/2023 CUST057 Female 63 Beauty 1 30 30
58 13/11/2023 CUST058 Male 18 Clothing 4 300 1200
59 5/7/2023 CUST059 Male 62 Clothing 1 50 50
60 23/10/2023 CUST060 Male 30 Beauty 3 50 150
61 9/4/2023 CUST061 Male 21 Beauty 4 50 200
62 27/12/2023 CUST062 Male 18 Beauty 2 50 100
63 5/2/2023 CUST063 Male 57 Electronics 2 25 50
64 24/1/2023 CUST064 Male 49 Clothing 4 25 100
65 5/12/2023 CUST065 Male 51 Electronics 4 500 2000
66 27/4/2023 CUST066 Female 45 Electronics 1 30 30
67 29/5/2023 CUST067 Female 48 Beauty 4 300 1200
68 10/2/2023 CUST068 Male 25 Electronics 1 300 300
69 30/4/2023 CUST069 Female 56 Beauty 3 25 75
70 21/2/2023 CUST070 Female 43 Clothing 1 300 300
71 14/7/2023 CUST071 Female 51 Beauty 4 25 100
72 23/5/2023 CUST072 Female 20 Electronics 4 500 2000
73 21/8/2023 CUST073 Male 29 Electronics 3 30 90
74 22/11/2023 CUST074 Female 18 Beauty 4 500 2000
75 6/7/2023 CUST075 Male 61 Beauty 4 50 200
76 25/3/2023 CUST076 Female 22 Electronics 2 50 100
77 9/7/2023 CUST077 Female 47 Clothing 2 50 100
78 1/7/2023 CUST078 Female 47 Clothing 3 500 1500
79 18/4/2023 CUST079 Male 34 Beauty 1 300 300
80 10/12/2023 CUST080 Female 64 Clothing 2 30 60
81 17/5/2023 CUST081 Male 40 Electronics 1 50 50
82 26/12/2023 CUST082 Female 32 Beauty 4 50 200
83 16/12/2023 CUST083 Male 54 Electronics 2 50 100
84 28/11/2023 CUST084 Female 38 Electronics 3 30 90
85 6/2/2023 CUST085 Male 31 Clothing 3 50 150
86 8/11/2023 CUST086 Male 19 Beauty 3 30 90
87 22/11/2023 CUST087 Female 28 Beauty 2 50 100
88 29/3/2023 CUST088 Male 56 Clothing 1 500 500
89 1/10/2023 CUST089 Female 55 Electronics 4 500 2000
90 6/5/2023 CUST090 Female 51 Electronics 1 30 30
91 25/3/2023 CUST091 Female 55 Electronics 1 500 500
92 25/8/2023 CUST092 Female 51 Electronics 4 30 120
93 14/7/2023 CUST093 Female 35 Beauty 4 500 2000
94 19/5/2023 CUST094 Female 47 Beauty 2 500 1000
95 24/11/2023 CUST095 Female 32 Clothing 2 30 60
96 19/12/2023 CUST096 Female 44 Clothing 2 300 600
97 13/10/2023 CUST097 Female 51 Beauty 2 500 1000
98 23/4/2023 CUST098 Female 55 Beauty 2 50 100
99 17/12/2023 CUST099 Female 50 Electronics 4 300 1200
100 16/6/2023 CUST100 Male 41 Electronics 1 30 30
Based on the relevant datasets you have chosen, you are required to perform at
least 4 (four) different types of descriptive data analytics (chapter 2, 3, 4, & 5)
and at least 2 (two) different types of predictive data analytics (chapter 6 & 7).

I) 4 DIFFERENT TYPES OF DESCRIPTIVE ANALYSIS

CHAPTER 2: ANALYTICS USING MICROSOFT EXCEL

AGE QUANTITY
Mean 40.03 Mean 2.51
Standard Error 1.34926 Standard Error 0.11146
Median 40.5 Median 2.5
Mode 51 Mode 2
Standard Deviation 13.49257559 Standard Deviation 1.114596
Sample Varience 182.049596 Sample Varience 1.242323
Kurtosis -1.11758103 Kurtosis -1.34636
Skewness -0.026004001 Skewness -0.00327
Range 46 Range 3
Minimum 18 Minimum 1
Maximum 64 Maximum 4
Sum Sum 251
Count 100 Count 100

PRICE PER UNIT TOTAL AMOUNT

Mean 186.7 Mean 476.2
Standard Error 19.20493 Standard Error 58.22796
Median 50 Median 135
Mode 50 Mode 100
Standard Deviation 192.0493 Standard Deviation 582.2796
Sample Varience 36882.94 Sample Varience 339049.56
Kurtosis -1.22806 Kurtosis 0.75759
Skewness 0.688026 Skewness 1.366316
Range 475 Range 1975
Minimum 25 Minimum 25
Maximum 500 Maximum 2000
Sum 18670 Sum 47620
Count 100 Count 100
 AGE QUANTITY PRICE PER UNIT TOTAL AMOUNT
Sum 251 18670 47620
Mean 40.03 2.51 186.7 476.2
Mode 51 2 50 100
Median 40.05 2.5 50 135
Range 46 3 475 1975
Variance 182.049596 1.2423232 36882.93939 339049.5556
Standard Deviation 13.49257559 1.114596 192.0493 582.2796
Coefficient of Varience (CV) 0.337061593 0.444062 1.028651928 1.222762746
Coefficient of Skewness (CS) -0.026004001 -0.00327 0.688026 1.366316
Coefficient of Kurtosis (CK) -1.11758103 -1.34636 -1.22806 0.75759
Maximum 64 4 500 2000
Minimum 18 1 25 25
1) Use IF Function to create a new column named AMOUNT SIZE. Amount more than
1000 quantities are named as Large Amount and orders less than 1000 is named as
Small Amount. Provide total number of LARGE dan SMALL.
2) Suppose with amount of at least 1000 are considered high demand, flag those high
demand orders as High using IF and AND function. Create a new column named HIGH
DEMAND. Provide number of high demands.
3) Use VLOOKUP and provide any random 4 customers to find values based on Gender,
Age, Product Category and Total Amount. You may use Data Validation to help you
finding the values.

4) Use INDEX to find values below:

CHAPTER 3 : DATA VISUALATION AND EXPLORATION

Sum of Sum of Sum of Price per Sum of Total

Row Labels Age Quantity Unit Amount
Female 2380 136 12305 31430
Beauty 942 54 5085 12685
Clothing 675 37 2815 5765
Electronics 763 45 4405 12980
Male 1623 115 6365 16190
Beauty 505 43 1140 2800
Clothing 524 39 2735 6340
Electronics 594 33 2490 7050
Grand Total 4003 251 18670 47620

QUANTITY OF PRODUCT CATEGORY

AND GENDER
60
50
40
30
54
20 45 43
37 39
33
10
0
Beauty Clothing Electronics Beauty Clothing Electronics
Female Male

This bar graph shows data from the retail sales of quantity and product category based on
gender. From the graph its clearly shows the quantity of beauty product is higher for female
than male which is 54 female compare to 43 male. For clothing category, male have more
numbers compare to female which is 39 and 37. For electronics category, female have 37 while
male have 33. We can conclude here that female bought more beauty and electronics product
than male while male bought more clothing than female.
 PRICE PER UNIT OF PRODUCT
CATEGORY AND GENDER
6000
5000
4000
3000
5085
2000 4405
2815 2735 2490
1000
1140
0
Beauty Clothing Electronics Beauty Clothing Electronics
Female Male

This bar graph shows data from the retail sales of price per unit and product category based on
gender. From this graph, female dominate the bar graph for price per unit for each category
with 5085 for beauty, 2815 for clothing and 4405 for electronics. Male only got 1140 for beauty,
2735 for clothing and 2490 for electronics. We can conclude female spend more than male for
this bar graph.

TOTAL AMOUNT PRICE OF PRODUCT

CATEGORY AND GENDER
14000
12000
10000
8000
6000 12685 12980

4000 7050
5765 6340
2000
2800
0
Beauty Clothing Electronics Beauty Clothing Electronics
Female Male

This bar graph shows data from the retail sales of total amount price and product category based
on gender. From this graph, female spend the highest for beauty with 12685 while male only
spend 2800 for beauty. For clothing, male have the highest which is 6340 compare to female
only 5765. Female spend the most at electronics with 12980 and male only spend less which is
7050. We can conclude that female spend the most at beauty and electronics and male spend
the most at clothing.

CHAPTER 4 (DESCRIPTIVE STATISTIC SUMMARY)

Quantity Price Per Unit

Mean 2.51 Mean 186.7

Standard Error 0.11146 Standard Error 19.20493
Median 2.5 Median 50
Mode 2 Mode 50
Standard Deviation 1.114596 Standard Deviation 192.0493
Sample Variance 1.242323 Sample Variance 36882.94
Kurtosis -1.34636 Kurtosis -1.22806
Skewness -0.00327 Skewness 0.688026
Range 3 Range 475
Minimum 1 Minimum 25
Maximum 4 Maximum 500
Sum 251 Sum 18670
Count 100 Count 100

PRICE PER
QUANTITY UNIT
Sum of Quantity 251 18670
No. of Observation 100 100
Mean quantity 2.51 186.7
Excel AVERAGE 3 475
Excel MEDIAN 2.5 50
Excel MODE 2 50
Range 3 475
Excel Variance 1.2423232 36882.93939
Excel Std Deviation 1.1145955 192.049315
COEFFICIENT OF VARIANCE (CV) 0.444062 1.028651928
COEFFICIENT OF SKEWNESS (CS) -0.003274 0.688026339
COEFFICIENT OF KURTOSIS (CK) -1.34636 -1.228063618

The coefficient of variation (CV) is the relative measure of dispersion that can be defined as
the ratio of standard deviation to the mean, which is CV = standard deviation/mean. It is often
expressed as a percentage by multiplying by 100. CV is used to determine the volatility of risk
to return, to compare the expected amount of return from two or more investments. The smaller
the CV, the better risk-return trade-off.

Based on the Retails Sales for 4 - Quarter of year 2023 database, the coefficients of variation
(CV) for Price per Unit were lower than Quantity.
CHAPTER 5
QUESTION

Discuss and interpret the result obtained from the above two analysis you have
chosen and provide some practical suggestions to the management to improve
their business performance

I) Interpret data from descriptive data analysis

• MEAN

The average, or mean, is the data's central tendency, a figure that represents the range
of the entire set of data. The mean is computed by dividing the total number of
observations by the data values.

The population of mean is denoted by µ=𝑥1 + 𝑥2 + 𝑥3 + …. + 𝑥𝑛 = ∑𝑁

𝑖=1 𝑥𝑖 where the

N is the number of population. N N

So for the mean of the data from the quantity is 2.51 and for the price is 186.7 so it’s
very useful to measure the data and to compare the differences between the data.

• MEDIAN

Median is a middle value in the data sets when the observations are arranged in
ascending orders. The values of the middle are the median if the total number of
observations is odd. The median, or average of the two middle values, is found in data
sets where the total number of values is even. The outliers have no effect on the middle
of the data. For data sets that include outliers, the median is the most accurate indicator
of central tendency.

So the formula is to add two middle numbers and divide by 2. So the median for the
data quantity is 2.5 and for the price is 50.

• STANDARD ERROR

The sample mean, or average, of the data is estimated to be a certain distance from the
genuine population mean using the standard error of the mean (SEM). Every time, the
SEM is less than the SD.

So the standard error for the quantity is 0.11146 and for the price is 19.20493.
• MODE

The data value that appears most frequently is the mode. However, if every value
appears the same number of times, a data collection may not have any mode at all. The
data set is bimodal if two values occur more frequently than the remaining values at the
same time. The data set is trimodal if three values emerged simultaneously and more
frequently than the remaining values; multimodal if n modes are present. So the mode
for the quantity is 2 and the mode for the price is 50.

• STANDARD DEVITIATION

The standard deviation between each item and the mean is measured using the standard
deviation. In other words, the dispersion of data from the mean. Whereas a high
standard deviation suggests that the data points are dispersed throughout a larger range
of values, a low standard deviation suggests that the data points often tend to be near
the data set mean.

So the formula for the population variance is σ = ∑ni=1(xi−¯x)2 / N

Meanwhile for the sample variance is is s = ∑ni=1(xi−¯x)2 / N

So the standard deviation for the quantity is 1.114596 and for the price is 192.04.

• SAMPLE VARIANCE

The square of the average distance between each quantity and the mean is the variance.
In other words, it is the standard deviation squared.

So the formula for the population variance is 𝜎 2 = ∑ni=1(xi−¯x)2 / N

Meanwhile for the sample variance is 𝑠 2 = ∑ni=1(xi−¯x)2 / N

So the sample variance for the quantity is 1.242323 and for the price is 36882.94.

• MINIMUM

Minimum is function to returns the smallest number in a set of values.

So the minimum values for the quantity is 1 and for the price is 25.
• MAXIMUM

Maximum is function to returns the highest number in a set of values.

So the maximum for the quantity is 4 and for the price is 500.

• KURTOSIS

A statistical metric called kurtosis is employed to characterize a feature of a dataset.

Plotting properly distributed data on a graph typically results in the shape of a bell. We
refer to this as the bell curve. The tails on either side of the curve are often formed by
the depicted data that deviate the most from the mean of the data. Kurtosis shows how
much information is in the tails.

So the kurtosis for the quantity is -1.34636 and for the price is -1.22806.

• SKEWNESS

Deviating from a symmetrical bell curve is known as skewness. It displays an

asymmetrical curve with a tail that extends either to the left (negative skew) or to the
right (positive skew). Skewness also refers to the fact that a data set's mean, median,
and mode are all distinct and do not share a common value.

So the Skewness for the quantity is -0.00327 and for the price is 0.688026.

• RANGE

One of the most basic methods in descriptive statistics is range. It is the variation
between the value that is lowest and highest.

So the range for the quantity is 3 and for the price is 475.

• SUM

Excel's SUM function adds the numbers within a range of cells. Enter the values to be
summed after typing "=SUM," as it falls within the Math and Trigonometry function.
The parameters passed to the function may consist of ranges, cell references, or numbers.

So the sum amount for the quantity is 251 and for the price is 18670.
 • COUNT

The count function counts the number of numbers in the list of inputs as well as the
number of cells that contain numbers. To find the number of entries in a number field
that are within a range or array of numbers, use the count function.

So the count amount for the quantity is 100 and for the price is 100.

II) Interpret data from predictive data analysis

III) Practical suggestions to improve the business performance

I. Set a key performance indicators (KPIs) and clear goals. The retails company should
be able to define goals that are SMART, which stands for specific, measurable,
attainable, relevant, and related to time. So, the company can easily to monitor growth
and success in important areas such as sales, customer satisfaction, and operational
efficiency, key performance indicators (KPIs) should be established.
II. Strategies successful marketing and branding strategies: Company need to establish a
powerful and consistent brand image that strikes a chord with the people you intend to
attract as customers. Company need to make use of digital marketing methods in order
to engage with customers and reach a larger audience.
III. Make investments in the training and development of your staff: Employees should be
provided with continual training in order to maintain their current skill set. It is
important to cultivate a culture within the organization that emphasizes innovation and
ongoing development.
CHAPTER 5: STATISTICAL INFERENCE

5.1 One-Tailed Test: Critical Value Approach

Step 1: Hypotheses

First of all, to in order to run this test, we decided to use one sample hypothesis which is the
one tailed test. The hypotheses testing that we decide are:

a. H0 = μ > 2. The average product quantity is greater than 2.505 per year.

b. H1 = μ < 2. The average product quantity is lesser than 2.505 per year.

Step 2: Significance level

Significance level, α and critical, t

Critical, t and Rejection area
α=the significance level 0.05
n-1= Degree of freedom 49
Critical, t = T.INV (α, n-1) 1.6895

The significance level given is 0.05 and t-distribution is applied. The < sign in alternative
hypothesis indicates that test is one-tailed with rejection area located at the lower/left tailed
on the t-distribution curve.

The critical t-distribution curve, t0.05, 49 is computed to 1.6895 with the formula of Critical t
=T.INV (α, n-1). As the rejection area lies in the right-tail, the critical t-distribution value is
positive.
Step 3: T-statistics

Critical Value (CV) 1.677

Mean 2.62
µ 2
std deviation 1.193349
n 50
t-stat 3.674

p-value 0.000296

T-statistics is applied by the formula of t =(x̅-μ)/s/sqrt of n. Hence the result of t-test is

amounted to 3.674.

Step 4: Decision rule

LOWER ONE TAILED T-STAT CRIT-T

i) if T-STAT < T-CRIT = REJECT H0 3.674 > 1.677
ii) if T-STAT > T-CRIT = FAIL TO REJECT
H0

Next, we compare the result of t test-statistics value with the critical value. Since the test
statistics value is 3.674 which is bigger than critical value of t = 1.677, it falls under non
rejection area. Therefore, we failed to reject H0. This means that H0 is supported.

Step 5: Conclusion

As a result of failed to reject H0, there is insufficient evidence to support the claim made by
H1 whereby the average product development is lesser than 2.505 per year. In conclusion, H0
is supported and the average product quantity development that has been stated is true which
is greater than 2.505 and the claim is not true.
5.2 Chi-Square Test for Independence

Step 1:

The hypotheses testing that we decide are:

a) H0 = Gender and product category are independent (or not related).

b) H1 = Gender and product category are dependent (or related).
Step 2:

The significance level that we decide is 0.05 and chi-square distribution is applied.

Step 3:

We have calculated the chi-square test for independence value using the excel functions
=CHISQ.INV.RT(α, deg_freedom) which is 5.991.

Step 3a:

Based on Excel Retails Sales for 4-quarter of year 2023 dataset, we have construct a cross-
tabulation table using PivotTable Function.
 Sum of total Amount Product Category
Gender Beauty Clothing Electronics Grand Total
Female 12685 5765 12980 31430

Male 2800 6340 7050 16190

Grand Total 15485 12105 20030 47620

Step 3b:

We have constructed the observed frequencies from the cross-tabulation.

Observed Frequency

Sum of total Amount Product Category

Gender Beauty Clothing Electronics Grand Total
Female 12685 5765 12980 31430

Male 2800 6340 7050 16190

Grand Total 15485 12105 20030 47620

Step 3c:

(𝑅𝑜𝑤 𝑇𝑜𝑡𝑎𝑙)(𝐶𝑜𝑙𝑢𝑚𝑛 𝑇𝑜𝑡𝑎𝑙

We calculated the expected frequencies using the formula, fe = in a
𝑛

new cross tabulation shown below.

Observed Frequencies

Sum of total Amount Product Category

Gender Beauty Clothing Electronics Grand Total
Female 12685 5765 12980 31430

Male 2800 6340 7050 16190

Grand Total 15485 12105 20030 47620

p-Value Method

Step 4:

We have calculate the p-value for one tailed hypothesis test with degree of freedom ( df = (R
- 1) (C -1) = 2 and using the excel functions = CHISQ.TEST(actual_range,expected_range) is
is 0.840.

df, degree of freedom 2

P-Value 0.840

Step 5:

We have concluded that the p-value of 0.840 is greater than the significance value, α = 0.05,
thus we fail to reject the null hypothesis. Therefore, we can conclude that gender and product
category are not related.
CHAPTER 6: TRENDLINES AND REGRESSION ANALYSIS

Retail Sales Per Month for Year 2023

MONTH PRODUCT QUANTITY OF TOTAL AMOUNT

PRODUCT SALES
JANUARY BEAUTY 4 590
CLOTHING 8 220
ELECTRONICS 7 3030
FEBRUARY BEAUTY 0 0
CLOTHING 17 4025
ELECTRONICS 10 1950
MARCH BEAUTY 5 500
CLOTHING 3 550
ELECTRONICS 5 700
APRIL BEAUTY 12 1205
CLOTHING 12 340
ELECTRONICS 3 80
MAY BEAUTY 11 2375
CLOTHING 4 1400
ELECTRONICS 13 4180
JUNE BEAUTY 4 950
CLOTHING 0 0
ELECTRONICS 5 1230
JULY BEAUTY 12 2300
CLOTHING 7 1950
ELECTRONICS 3 130
AUGUST BEAUTY 8 1025
CLOTHING 0 0
ELECTRONICS 11 1740
SEPTEMBER BEAUTY 0 0
CLOTHING 2 50
ELECTRONICS 0 0
 OCTOBER BEAUTY 18 2320
CLOTHING 6 300
ELECTRONICS 6 3000
NOVEMBER BEAUTY 16 3870
CLOTHING 10 2460
ELECTRONICS 3 90
DECEMBER BEAUTY 7 350
CLOTHING 7 810
ELECTRONICS 12 3900

Scatter Plot of Retails Sales per Month for Year 2023 and The Best-Fitted Line

Retails Sales Per Month For Year 2023

4500

4000

3500

3000
Total Amount Sales

2500

2000

1500

1000

500

0
0 2 4 6 8 10 12 14 16 18 20
-500
Quantity Of Product

Interpretation: In the scatter plot, we observe a positive relationship between the Total
amount (x-axis) and The Quantity of Product (y-axis). As the total amount increases, the
quantity of the product tends to rise. The fitted linear function represents the best line through
data points, It has a positive slope, indicating that, on average, the total amount increases for
every quantity of product sold. The scatter plot points are clustered significantly close to the
fitted line, indicating a strong linear connection. This argument implies that the Retail Sales
per Month for the Year 2023 l approximates the underlying trend well.
Simple Linear Regression Model

Interpreting Regression Analysis for Simple Linear Regression Model

Analysis Interpretation
Multiple R 0.763 < 0- negative correlation

R-Squares R² = 0. 58
It means that approximately 58% of the variation in total amount
sales is explained by the quantity product.

Standard Error The standard error is 3.3 which is small thus, the data is clustered.

Anova p-value for the F-test = 0.00000006

is less than the level of significance (0.05).
Therefore, the null hypothesis is rejected. This means that the
selling price is a statistically significant variable in explaining the
variation in the Retail sales for the year 2023.
Testing hypothesis for t = 0.00286 / 0.00042
regression coefficient
t-test t = 6.809

The p-values for both the coefficients are almost zero; therefore,
we can conclude that both the coefficients are statistically not
equal to zero (reject Ho).
For a simple linear regression (one independent variable), the p-
value for the t-test and the F-test will always be similar.
 CHAPTER 7: FORECASTING

7.1 Forecasting Time Series Using Moving Average

month total 0.1 ad 0.3 ad 0.6 ad 0.9 ad
amount
January 3840.00 #N/A #N/A #N/A #N/A
February 5975.00 3840.00 2135.00 3840.00 2135.00 3840.00 2135.00 3840.00 2135.00
March 1750.00 4053.50 2303.50 4480.50 2730.50 5121.00 3371.00 5761.50 4011.50
April 1625.00 3823.15 2198.15 3661.35 2036.35 3098.40 1473.40 2151.15 526.15
May 7955.00 3603.34 4351.67 3050.45 4904.56 2214.36 5740.64 1677.62 6277.39
June 2180.00 4038.50 1858.50 4521.81 2341.81 5658.74 3478.74 7327.26 5147.26
July 4380.00 3852.65 527.35 3819.27 560.73 3571.50 808.50 2694.73 1685.27
August 2765.00 3905.39 1140.39 3987.49 1222.49 4056.60 1291.60 4211.47 1446.47
September 50.00 3791.35 3741.35 3620.74 3570.74 3281.64 3231.64 2909.65 2859.65
October 5620.00 3417.21 2202.79 2549.52 3070.48 1342.66 4277.34 335.96 5284.04
November 6420.00 3637.49 2782.51 3470.66 2949.34 3909.06 2510.94 5091.60 1328.40
December 5060.00 3915.74 1144.26 4355.46 704.54 5415.62 355.62 6287.16 1227.16

MAD 2216.86 2384.23 2606.77 2902.57

Conclusion: By comparing MAD, exponential smoothing model with α=0.1 is the lowest compared to
others with 2216.86 which is the perfect forecast.