In: Statistics and Probability
Fill in all the underlined spots on the spreadsheet with the data about Absorbance of Light for different Nitrate Levels. The goals are: 1) to compute the correlation and slope of the regression line by using the "SS formulas" and 2) to compute SSE, the sum of the squared "errors" (residuals).
Data from Exercise 2.69 (p. 97) | ||||||||
The absorbance of Light for Different Nitrate Levels | ||||||||
y - that | ||||||||
Nitrates x (mg/liter of water) | Absorbance y | x^2 values | y^2 values | x*y values | y hat (predicted absorbances) | residuals/errors | squared residuals | |
50 | 7 | _____ | _____ | _____ | _____ | _____ | _____ | |
50 | 7.5 | _____ | _____ | _____ | _____ | _____ | _____ | |
100 | 12.8 | _____ | _____ | _____ | _____ | _____ | _____ | |
200 | 24 | _____ | _____ | _____ | _____ | _____ | _____ | |
400 | 47 | _____ | _____ | _____ | _____ | _____ | _____ | |
800 | 93 | _____ | _____ | _____ | _____ | _____ | _____ | |
1200 | 138 | _____ | _____ | _____ | _____ | _____ | _____ | |
1600 | 183 | _____ | _____ | _____ | _____ | _____ | _____ | |
2000 | 230 | _____ | _____ | _____ | _____ | _____ | _____ | |
2000 | 226 | _____ | _____ | _____ | _____ | _____ | _____ | |
Sums | 8400 | 968.3 | _____ | _____ | _____ | SSE | _____ | |
Means | 840 | 96.83 | ||||||
Std Devs | 802.7037644 | 90.95273559 | ||||||
Correlation | 0.999939232 | |||||||
Coefficient of Determination | 0.999878467 | SSxx | _____ | |||||
SSyy | _____ | |||||||
Slope | 0.113301086 | SSxy | _____ | |||||
Intercept | 1.657087429 | |||||||
Correlation | _____ | |||||||
Regression Equation | y = 0.1133x + 1.6571 | Slope | _____ |
Completed Table
Nitrates x (mg/liter of water) |
Absorbance y |
x^2 values |
y^2 values |
x*y values |
y hat (predicted absorbances) |
residuals/errors |
squared residuals |
|
50 |
7 |
2,500 |
49 |
350 |
7.32 |
-0.32 |
0.10 |
|
50 |
8 |
2,500 |
56 |
375 |
7.32 |
0.18 |
0.03 |
|
100 |
13 |
10,000 |
164 |
1,280 |
12.99 |
-0.19 |
0.04 |
|
200 |
24 |
40,000 |
576 |
4,800 |
24.32 |
-0.32 |
0.10 |
|
400 |
47 |
1,60,000 |
2,209 |
18,800 |
46.98 |
0.02 |
0.00 |
|
800 |
93 |
6,40,000 |
8,649 |
74,400 |
92.30 |
0.70 |
0.49 |
|
1,200 |
138 |
14,40,000 |
19,044 |
1,65,600 |
137.62 |
0.38 |
0.15 |
|
1,600 |
183 |
25,60,000 |
33,489 |
2,92,800 |
182.94 |
0.06 |
0.00 |
|
2,000 |
230 |
40,00,000 |
52,900 |
4,60,000 |
228.26 |
1.74 |
3.03 |
|
2,000 |
226 |
40,00,000 |
51,076 |
4,52,000 |
228.26 |
-2.26 |
5.10 |
|
Total |
8,400 |
968 |
128,55,000 |
1,68,212 |
14,70,405 |
9.05 |
SSE = ∑Squared residuals = 9.05
SSX = ∑X2 – (∑X)2/n = 128,55,000 – (8400)2/10 = 57,99,000
SSY = ∑Y2 – (∑Y)2/n = 168212 – (968)2/10 = 74,452
SSXY = ∑XY – (∑X*∑Y)/n = 1470405 – (8400*968)/10 = 6,57,033
SSReg = (SSXY)2/SSX = 74442.55
SSTotal = SSY
Correlation Coefficient = (SSReg / SSTotal)1/2 = 0.999939
Slope = SSXY/SSX = 6,57,033/57,99,000 = 0.113301