Skip to main content
Chemistry LibreTexts

Study Session 2: Probability and Statistics

[ "article:topic", "showtoc:no" ]
  • Page ID
  • The uncertain nature of things in the quantum world make it necessary for us to use probability and statistics to describe the likelihood of finding a system in a certain state. That means a short review is useful.

    Flip a coin 20 times and record the outcome. How many times did your coin flip result in heads? How often do you expect to get heads?


    If you roll a six-sided die 60 times, how many times do you expect your roll to result in? Why?


    If \(N_j\) is the number of times that you measure outcome \(j\), out of \(N\) total repetitions, then

    \(p_j=\lim_{N\rightarrow \infty}\frac{N_j}{N}\)

    What values can \(p_j\) have? What does it mean if \(p_j=1\)? what deos it mean if \(p_j=0\)?


    If \(N_j\) has \(n\) possible values, what is the value of \(\sum_{j=1}^{n}N_j\)?


    What is the value of \(\sum_{j=1}^{n}p_j\)?


    Using \(p_j\), how would you describe the average value of your die roll?


    We can write the average or mean of a measurement using

    \(\left \langle x \right \rangle =\sum_{j=1}^{n}x_jp_j=\sum_{j=1}^{n}x_jp(x_j)\)

    What is the probability of rolling any particular value 1-6 on a die? What is the average value that you will roll? 


    In addition to the mean or average, we can compute the second moment, \( \left \langle x^2 \right \rangle = \sum_{j=1}^{n}x_j^2p_j\) and with this, we define the variance

    \(\sigma_x^2=\left \langle (x-\left \langle x \right \rangle )^2 \right \rangle = \sum_{j=1}^{n}(x-\left \langle x \right \rangle )^2p_j\)

    How do you get a large value for the variance? How do you get a small value for the variance? What does the variance measure?


    Complete the square in the sum for the variance and simplify the expression.



    What happens when our distribution is continuous instead of discrete? Then the probability of finding the system in an infinitesimal position between \(x\) and \(x+dx\) is \(p(x)dx\). 


    How do we express the probability of finding a particle between \(x=a\) and \(x=b\)?


    The most common distribution we'll use is a Gaussian, \(p(x)dx=ce^{-x^2/2a^2}dx\) for \(-\infty < x < \infty \). 

    Find \(c, \left \langle x \right \rangle , \sigma_x^2\), and \(\sigma_x\)