In: Statistics and Probability
Mist (airborne droplets or aerosols) is generated when metal-removing fluids are used in machining operations to cool and lubricate the tool and workpiece. Mist generation is a concern to OSHA, which has recently lowered substantially the workplace standard. An article gave the accompanying data on x = fluid-flow velocity for a 5% soluble oil (cm/sec) and y = the extent of mist droplets having diameters smaller than 10 µm (mg/m3):
x | 88 | 177 | 182 | 354 | 369 | 442 | 961 |
y | 0.41 | 0.60 | 0.46 | 0.66 | 0.58 | 0.69 | 0.96 |
(b) What proportion of observed variation in mist can be attributed to the simple linear regression relationship between velocity and mist? (Round your answer to three decimal places.)
answer= 0.908
(c) Calculate the test statistic and determine the P-value. (Round your test statistic to two decimal places and your P-value to three decimal places.)
H0: β1 = 0.0006667
Ha: β1 < 0.0006667
t=______________
P-value=_____________
(d) Estimate the true average change in mist associated with a 1 cm/sec increase in velocity, and do so in a way that conveys information about precision and reliability. (Calculate a 95% CI. Round your answers to six decimal places.)( , ) mg/m3
X | Y | XY | X² | Y² |
88 | 0.41 | 36.08 | 7744 | 0.1681 |
177 | 0.6 | 106.2 | 31329 | 0.36 |
182 | 0.46 | 83.72 | 33124 | 0.2116 |
354 | 0.66 | 233.64 | 125316 | 0.4356 |
369 | 0.58 | 214.02 | 136161 | 0.3364 |
442 | 0.69 | 304.98 | 195364 | 0.4761 |
961 | 0.96 | 922.56 | 923521 | 0.9216 |
Ʃx = | Ʃy = | Ʃxy = | Ʃx² = | Ʃy² = |
2573 | 4.36 | 1901.2 | 1452559 | 2.9094 |
Sample size, n = | 7 |
x̅ = Ʃx/n = 2573/7 = | 367.5714286 |
y̅ = Ʃy/n = 4.36/7 = | 0.622857143 |
SSxx = Ʃx² - (Ʃx)²/n = 1452559 - (2573)²/7 = | 506797.7143 |
SSyy = Ʃy² - (Ʃy)²/n = 2.9094 - (4.36)²/7 = | 0.193742857 |
SSxy = Ʃxy - (Ʃx)(Ʃy)/n = 1901.2 - (2573)(4.36)/7 = | 298.5885714 |
b) Coefficient of determination, r² = (SSxy)²/(SSxx*SSyy)
= (298.58857)²/(506797.71429*0.19374) = 0.9080
c)
Null and alternative hypothesis:
Ho: β₁ = 0.0006667
Ha: β₁ < 0.0006667
n = 7
α = 0.05
Slope, b = SSxy/SSxx = 298.58857/506797.71429 = 0.000589167
Sum of Square error, SSE = SSyy -SSxy²/SSxx
= 0.19374 - (298.58857)²/506797.71429 = 0.017824276
Standard error, se = √(SSE/(n-2)) = √(0.01782/(7-2)) = 0.05971
Test statistic:
t = (b-β₁)/(se/√SSxx) = (0.000589 - 0.0006667)/(0.05971/√506797.7143) = -0.92
df = n-2 = 5
p-value = T.DIST.2T(ABS(-0.9244), 5) = 0.199
Conclusion:
p-value > α , Fail to reject the null hypothesis.
d) Critical value, t_c = T.INV.2T(0.05, 5) = 2.5706
95% Confidence interval for slope:
Lower limit = β₁ - tc*se/√SSxx = 0.0006 - 2.5706*0.0597/√506797.7143 = 0.000374
Upper limit = β₁ + tc*se/√SSxx = 0.0006 + 2.5706*0.0597/√506797.7143 = 0.000805