1. Some probability density functions (pdfs) Normal: outcome influenced by a very large number of very small, ‘50-50 chance’ effects (Ex: human heights) Lognormal: outcome influenced by a very large number of very small, ‘constrained’ effects (Ex: rain drops) Poisson: outcome influenced by a rarely occurring events in a very large population (Ex: micrometeoroid diameters in LEO) Weibull: outcome influenced by a ‘failure’ event in a very large population (Ex: component life time) Binomial: outcome influenced by a finite number of ’50-50 chance’ effects (Ex: coin toss)

2. Figure 8.4 The Normal Distribution

3. Concept of the Normal Distribution Population, with true mean and true variance x‘ and σ Figure 8.5 Experiment with many, small, uncontrolled extraneous variables

4. Normalized Variables β = (x-x') / σstandardized normal variable z 1 = (x 1 -x') / σnormalized z-variable (z 1 is a specific value of β); subscript 1 usually dropped where p(z 1 ) is the normal error function.

5. Table 8.2 Normal Error Function Table Pr[0≤z≤1] = Pr[-1≤z≤1] = Pr[-2≤z≤2] = Pr[-3≤z≤3] = 0.6826 Pr[-0.44≤z≤4.06] = 0.3413 0.9544 0.9974 0.1700+0.5000 =0.6700 or 67 %

6. In-Class Problem What is the probability that a student will score between 75 and 90 on an exam, assuming that the scores are distributed normally with a mean of 60 and a standard deviation of 15 ? z 75 = (75-60)/15 = 1 and z 90 = (90-60)/15 = 2 one-sided z-table >> P 75 = 0.3413 and P 90 = 0.4772 >> P 75 to 90 = 0.4772 – 0.3413 = 0.1359

7. Statistics Using LabVIEW Another name for the probability distribution function (PDF) is the cumulative distribution function (CDF or cdf). Why is it called cumulative? Here is where you can find CDF functions and the inverses of these functions in LabVIEW Let’s solve the previous problem using LabVIEW