Annual rainfall data for the Alvin, Texas gage are given below. The data should be fitted using a log Pearson type 3 distribution. Decide if 1979 is an outlier by performing the analysis with and without the data point included.

To determine if a value is an outlier, perform the following analysis, as presented by the Interagency Committee on Water Data (1982): Determine the high and low outlier thresholds of the distribution. If an outlier occurs, then discard it from the data set and repeat the analysis. These values can be calculated from the following equations:
yH = μ + Knσ
yL = μ - Knσ
Where Kn is the one-sided 10% significance level for the normal distribution, a function of n. (For n = 21, Kn = 2.408. For n = 20, Kn = 2.385 ) The value yH is the high outlier threshold (in log units for the lognormal or LP3 distributions), yL is the low outlier threshold (in log units for the lognormal or LP3 distributions), μ is the mean (of the log- transformed data for the lognormal or LP3 distributions), and σ is the standard deviation. Using log-converted data perform a Log-Pearson analysis with and without the 1979 data point.

Transcribed Image Text:

5 it 5 9 1 455 789 e9 3385998 5451784 412122 340454 7 8 9 0 1 2 3 78888 3532 9875 4357, 5 4 5 0123456 e-999