This preview has intentionally blurred sections. Sign up to view the full version.
View Full DocumentThis preview has intentionally blurred sections. Sign up to view the full version.
View Full Document
Unformatted text preview: Problem 2.52. (a) A normal quantile plot of x follows. Clearly the variable, hourly median power, is not normally distributed, as the normal quantile plot is curvilinear.21 1 2 50 100 150 200 250 z quantile power b) By taking the natural logarithm of the variable and constructing a normal quantile plot of it, we obtain the plot above. This plot looks quite linear indicating that it is plausible that these observations were sampled from a lognormal distribution. Problem 3.4 (a) 150 100 50 BOD mass loading 90 80 70 60 50 40 30 20 10 BOD mass removal21 1 2 1 2 3 4 5 6 z quantile ln(power) On both the BOD mass loading boxplot and the BOD mass removal boxplot there are 2 outliers. Both variables are positively skewed. (b) There is a strong linear relationship between BOD mass loading and BOD mass removal. As the BOD mass loading increases so does the BOD mass removal. The two outliers seen on each of the boxplots are seen to be correlated here. There is one observation that appears not to match the linear pattern. This value is (37, 9). One might have expected a larger value for BOD mass removal....
View
Full Document
 Spring '08
 marzban
 Least Squares, Regression Analysis, 96.1%, BOD mass removal, BOD mass

Click to edit the document details