De La Salle University – Dasmariñas

COLLEGE OF SCIENCE AND COMPUTER STUDIES

MATHEMATICS AND STATISTICS DEPARTMENT
City of Dasmariñas, Cavite

SMATH001LA – Data Analytics for Engineering

2nd Semester / Midterm Period / S.Y. 2020-2021

LABORATORY ACTIVITY #6
CORRELATION AND REGRESSION ANALYSIS

Score:
NAME: ISAIAH JOHN DOMENIC M. CANTANERO DATE: JANUARY 14, 2022

COURSE/YEAR & SECTION: BSA24 PROF.: SIR DHENMAR CHUA

All Greens Franchise

The data (X1, X2, X3, X4, X5, X6) are for each franchise store.
X1 = annual net sales/$1000
X2 = area of store in sq. ft./1000
X3 = inventory/$1000
X4 = amount spent on advertizing/$1000
X5 = size of sales district/1000 families
X6 = number of competing stores in district

Perform the following:

1. Determine which variables are correlated.

Pearson's Correlations

Pearson's r p Interpretation
<. Very Strong Positive
X1 - X2 0.894 ***
001 Correlation
<. Very Strong Positive
X1 - X3 0.946 ***
001 Correlation
<. Very Strong Positive
X1 - X4 0.914 ***
001 Correlation
<. Very Strong Positive
X1 - X5 0.954 ***
001 Correlation
X1 - X6 - *** <. Very Strong Negative
 0.912 001 Correlation
<. Very Strong Positive
X2 - X3 0.844 ***
001 Correlation
X2 - X4 0.749 *** <. Strong Positive Correlation
001
<. Very Strong Positive
X2 - X5 0.838 ***
001 Correlation
- <.
X2 - X6 ***
0.766 001 Strong Negative Correlation
<. Very Strong Positive
X3 - X4 0.906 ***
001 Correlation
<. Very Strong Positive
X3 - X5 0.864 ***
001 Correlation
- <. Very Strong Negative
X3 - X6 ***
0.807 001 Correlation
X4 - X5 0.795 *** <. Strong Positive Correlation
001
- <. Very Strong Negative
X4 - X6 ***
0.841 001 Correlation
<. Very Strong Negative
X5 - X6 -0.87 ***
001 Correlation

* p < .05, ** p < .01, *** p <

.001
The table shows the Pearson’s correlation of the given data which mostly results to Positive
Correlation.

Linear Regression
Model Summary - annual net sales
Durbin-Watson
Mode Autocorrelatio
R R² Adjusted R² RMSE Statistic p
l n
192.06
H₀ 0.000 0.000 0.000 0.252 1.479 0.164
2
H₁ 0.894 0.799 0.791 87.725 -0.016 1.438 0.115
 
ANOVA
Model   Sum of Squares df Mean Square F p
H₁ Regression 766689.456 1 766689.456 99.627 < .001
  Residual 192390.896 25 7695.636  
  Total 959080.352 26  
Note.  The intercept model is omitted, as no meaningful information can be shown.
 
Coefficients
95% CI
Unstandardize Standard Standardize
Model   t p Lower Upper
d Error d
< .00
H₀ (Intercept) 286.574 36.962 7.753 210.597 362.551
1
H₁ (Intercept) 2.577 33.085 0.078 0.939 -65.562 70.716
area of store in < .00
  85.389 8.555 0.894 9.981 67.770 103.008
sq. ft. 1
 
Descriptives
  N Mean SD SE
192.06 36.96
annual net sales 27 286.574
2 2
area of store in sq.
27 3.326 2.011 0.387
ft.

 The R2 reveals that the store's area explains

79.9% of the variation in annual net sales, with the remaining 20.1 percent explained by other variables.
 The ANOVA findings demonstrate that we have a well-fitting model. The intercept 2.577 is not
significant in the coefficients section because its p-value is more than 0.5, but the 85.389 is.
 Annual Net Sales = 2.577 + 85.389*Store Area in sq ft would be the formula.
Linear Regression

Model Summary - annual net sales

Durbin-Watson
Mode Autocorrelatio
R R² Adjusted R² RMSE Statistic p
l n
192.06
H₀ 0.000 0.000 0.000 0.252 1.479 0.164
2
H₁ 0.946 0.894 0.890 63.776 0.219 1.550 0.210
 
ANOVA
Model   Sum of Squares df Mean Square F p
H₁ Regression 857395.877 1 857395.877 210.798 < .001
  Residual 101684.475 25 4067.379  
  Total 959080.352 26  
Note.  The intercept model is omitted, as no meaningful information can be shown.
 
Coefficients
95% CI
Mode Standardize
  Unstandardized Standard Error t p Lower Upper
l d
< .00 362.55
H₀ (Intercept) 286.574 36.962 7.753 210.597
1 1
H₁ (Intercept) -81.504 28.167 -2.894 0.008 -139.514 -23.494
14.51 < .00
  inventory 0.950 0.065 0.946 0.815 1.085
9 1
 
Descriptives
  N Mean SD SE
annual net 36.96
27 286.574 192.062
sales 2
36.79
inventory 27 387.481 191.168
0

 The R2 reveals that inventory accounts for 89.4 percent of the variation in annual net sales, with other
variables accounting for the remaining 10.4 percent.
 The ANOVA findings demonstrate that we have a well-fitting model. Because the p-value is less than
0.05, the coefficients -81.504 and 0.950 are significant.
 Annual Net Sales = -81.504 + 0.950*Inventory is the formula.
Linear Regression

Model Summary - annual net sales

Mode
R R² Adjusted R² RMSE
l
H₀ 0.000 0.000 0.000 192.062
H₁ 0.914 0.835 0.829 79.455
 
ANOVA
Model   Sum of Squares df Mean Square F p
H₁ Regression 801254.098 1 801254.098 126.920 < .001
  Residual 157826.254 25 6313.050  
  Total 959080.352 26  
Note.  The intercept model is omitted, as no meaningful information can be shown.
 
Coefficients
Mode Unstandardize Standardize
  Standard Error t p
l d d
< .00
H₀ (Intercept) 286.574 36.962 7.753
1
H₁ (Intercept) -90.150 36.770 -2.452 0.022
amount spent on 11.26 < .00
  46.509 4.128 0.914
advertizing 6 1

 The R2 suggests that the amount spent on advertising accounts for 83.5 percent of the variation in
annual sales, with the remaining 16.5 percent explained by other variables.
 The ANOVA findings indicate that our model is well-fitting. The -90.150 and 46.509 coefficients are
statistically significant.
 Annual Net Sales = -90.150 + 46.509*Amount Spent on Advertising would be the formula.
Linear Regression

Model Summary - annual net sales

Durbin-Watson
Mode Autocorrelatio
R R² Adjusted R² RMSE Statistic p
l n
192.06
H₀ 0.000 0.000 0.000 0.252 1.479 0.164
2
H₁ 0.954 0.910 0.906 58.919 -0.273 2.535 0.171
 
ANOVA
Model   Sum of Squares df Mean Square F p
H₁ Regression 872294.494 1 872294.494 251.278 < .001
  Residual 86785.858 25 3471.434  
  Total 959080.352 26  
Note.  The intercept model is omitted, as no meaningful information can be shown.
 
Coefficients
95% CI
Mode Unstandardize Standard Standardize
  t p Lower Upper
l d Error d
< .00
H₀ (Intercept) 286.574 36.962 7.753 210.597 362.551
1
-
H₁ (Intercept) -58.823 24.563 -2.395 0.024 -8.235
109.412
size of sales < .00
  35.635 2.248 0.954 15.852 31.005 40.265
district 1
 
Descriptives
  N Mean SD SE
192.06
annual net sales 27 286.574 36.962
2
size of sales district 27 9.693 5.140 0.989

 The R2 demonstrates that the size of the sales district explains 91% of the variation in annual net sales,
with the remaining 9% explained by other variables.
 The ANOVA findings indicate that our model is well-fitting. Statistically significant coefficients are
-58.823 and 35.635.
 Annual Net Sales = -58.823 + 35.635*Size of Sales District would be the formula.
Linear Regression

Model Summary - annual net sales

Durbin-Watson
Mode Autocorrelatio
R R² Adjusted R² RMSE Statistic p
l n
192.06
H₀ 0.000 0.000 0.000 0.252 1.479 0.164
2
H₁ 0.912 0.832 0.825 80.239 0.151 1.581 0.256
 
ANOVA
Model   Sum of Squares df Mean Square F p
H₁ Regression 798122.917 1 798122.917 123.965 < .001
  Residual 160957.435 25 6438.297  
  Total 959080.352 26  
Note.  The intercept model is omitted, as no meaningful information can be shown.
 
Coefficients
95% CI
Unstandardize Standard Standardize
Model   t p Lower Upper
d Error d
< .00
H₀ (Intercept) 286.574 36.962 7.753 210.597 362.551
1
< .00
H₁ (Intercept) 563.593 29.283 19.246 503.283 623.902
1
number of
- < .00
  competing stores -35.787 3.214 -0.912 -42.407 -29.167
11.134 1
in district
 
Descriptives
  N Mean SD SE
2 36.96
annual net sales 286.574 192.062
7 2
number of competing stores in 2
7.741 4.896 0.942
district 7

 The R2 demonstrates that the number of competing stores in the district explains 83.2 percent of the
variation in annual net sales, with the remaining 16.8 percent explained by other variables.
 The ANOVA findings indicate that our model is well-fitting. 563.593 and -35.787 are statistically
significant intercepts.
 Annual Net Sales = 563.593 – 35.787*Number of Competing Stores in the District is the formula.